Model Host: Airbnb Price Predictions for hosts off the beaten path

Amsterdam Apartment, iamsterdam.com

Presentation - Youtube Link

Please refer to our presentation video for executive summary of this project/app!

https://youtu.be/Xr2tM-RtyQQ

Background Information – Amsterdam Airbnb Market

Amsterdam as a world-class tourism city greets millions of tourists from all over the world. However, like many other European cities, it’s so popular during the Summer that city officials want to decrease their visitors to improve the experience for all tourists and residents. Those who live in Amsterdam permanently not only face frequent belligerently intoxicated tourists on ‘moral holidays’, but they also pay more for housing – perhaps due to the short-term rental (STR) market propped up by the crush of visitors. Regulations have been passed to limit the spread and availability of STR’s like Airbnb through local bans and annual day limits on full home rentals, but their execution upsets many landlords. Owners want more consistency between cities and less fragility in their investments, but the regulations will continue to change as policymakers reach a consensus.

Airbnb’s are currently clustered in Amsterdam toward the historic centre of the city, but the market may shift. Crowd control and rental regulations closer to the center may result in more opportunities for those outside the central city to host B&B’s, private rooms, or an STR during their vacation elsewhere. With a new host market and with more data coming in from local STR registration requirements, data professionals can create a better B&B set up process through tools like price prediction.

Like any business or landlord, Airbnb hosts are motivated to list their property by the potential profit. Price prediction serves new and current hosts by giving them a better estimation of their bottom line than the usual set up process. Airbnb itself gives general guidelines and offers a model of their own to dynamically adjust pricing, but it still requires a maximum and minimum. Other tools display benchmarks for hosts, but they have fewer listings to compare outside the cluster. As data professionals, we can do this better.

Model Host, the name we’re giving our prediction tool, can serve these new people, especially ones outside the historic centre, with estimates of a current or potential property’s rental value. Anticipating Amsterdam’s mission to trade high-quantity tourism for high-quality experiences, our model here aims to grow responsible hosting opportunities, and promote a diversity of Airbnb locations in Amsterdam. When integrated into a user interface, residents could use price prediction to quickly see what opportunities are available to them. The following analysis aims to build this tool to benefit hosts and their visitors.

Data Source

We acquired data from several reliable sources:

Airbnb Data (including price, location, amenities, room compositions, etc.) is acquired from Kaggle, a platform for sharing data set particularly for machine learning. This contains airbnb data in Amsterdam from 2017 December to 2018 December (hence, primarily data for the year of 2018). Although due to the pandemic, tourism in Amsterdam might have undergone changes and variation, this is the best possible data we have available, and we believe it would produce some insights. (https://www.kaggle.com/datasets/erikbruin/airbnb-amsterdam?select=listings.csv)

Neighborhood and surrounding amenities data is acquired from City of Amsterdam’s online geo data portal, ‘Maps Data.’ (https://maps.amsterdam.nl/open_geodata/?LANG=en)

In addition, we supplemented data about tourist attrations from ‘Tourpedia,’ an online data portal with tourism-related data of several European cities. (http://tour-pedia.org/about/datasets.html)

source("https://raw.githubusercontent.com/urbanSpatial/Public-Policy-Analytics-Landing/master/functions.r")

library(tidyverse)
library(sf)
library(RSocrata)
library(viridis)
library(spatstat)
library(ggplot2)
library(raster)
library(spdep)
library(FNN)
library(mapview)
library(grid)
library(gridExtra)
library(knitr)
library(stringr)
library(kableExtra)
library(tidycensus)
library(lubridate)
library(viridis)
library(stargazer)
library(dplyr)
library(plyr)

library(scales)
library(RColorBrewer)
library(gridExtra)
library(ggthemes)
library(readr)
library(ggcorrplot)
library(caret)

library(sjPlot)
library(sjmisc)
library(sjlabelled)
library(osmdata)

#text mining
library(tm)
library(wordcloud2)
library(SnowballC)
options(scipen=999)

details.sf <- st_as_sf(details,coords = c('longitude','latitude'),crs = 4326) %>% 
  st_transform(st_crs(neighborhood))

details.sf$price <- parse_number(details.sf$price)
details.sf$weekly_price <- parse_number(details.sf$weekly_price)
details.sf$monthly_price <- parse_number(details.sf$monthly_price)
details.sf$cleaning_fee <- parse_number(details.sf$cleaning_fee)
details.sf$extra_people <- parse_number(details.sf$extra_people)
details.sf$security_deposit <- parse_number(details.sf$security_deposit)
details.sf$beds <- as.numeric(details.sf$beds)
details.sf$minimum_nights <- as.numeric(details.sf$minimum_nights)
details.sf$maximum_nights <- as.numeric(details.sf$maximum_nights)
details.sf$number_of_reviews <- as.numeric(details.sf$number_of_reviews)
details.sf$review_scores_rating <- as.numeric(details.sf$review_scores_rating)
details.sf$review_scores_accuracy <- as.numeric(details.sf$review_scores_accuracy)
details.sf$review_scores_cleanliness<- as.numeric(details.sf$review_scores_cleanliness)
details.sf$review_scores_value <- as.numeric(details.sf$review_scores_value)
details.sf$reviews_per_month <- as.numeric(details.sf$reviews_per_month)
details.sf$id <- as.character(details.sf$id)
details.sf$property_type <- as.factor(details.sf$property_type)

details.sf.raw <- details.sf

Exploratory Analysis

Since the Airbnb data already contains a lot of information (hiding many secrets!), we will first conduct exploratory analysis by examining location (geographic distribution) of available units, price per night, occupancy, room characteristics (composition, layout), and amenities that come with the room. Looking into those aspects can help our clients, future Airbnb host, to quickly identify internal characteristics of their unit and set price. It may also help those hosts who wish to upgrade their units figure out a potential increase in price after the upgrade.

# Available Calendar
available_calendar <- calendar %>%
  filter(available =="t")

available_calendar$listing_id <- as.character(available_calendar$listing_id)

# Price change
available_calendar <- available_calendar%>%
  mutate(price2 = gsub("^.","",price))%>%
  mutate(price2 = gsub(",","",price2))

available_calendar$price3 <- as.numeric(available_calendar$price2)

# Sum per month
available_calendar2 <- available_calendar%>%
  mutate(date2 = ymd(date))%>%
  mutate(month = month(date2))%>%
  group_by(listing_id, month) %>%
  dplyr::summarize(month_price = mean(price3))

length(unique(calendar$listing_id))

## [1] 20030

length(unique(calendar$listing_id))*12

## [1] 240360

study.panel <- 
  expand.grid(listing_id = unique(calendar$listing_id),
              month=unique(available_calendar2$month))
study.panel$listing_id <- as.character(study.panel$listing_id)
listing_panel <- study.panel %>%
  left_join(available_calendar2) %>%
  mutate(each_month_price = 0)
o <- order(listing_panel[,"listing_id"],listing_panel[,"month"])
listing_panel <- listing_panel[o,]
xx <- 0
for(i in 2:nrow(listing_panel)){
  if(!is.na(listing_panel[i,3])){
    xx <- listing_panel[i,3]
    listing_panel[i,4]=listing_panel[i,3]
  }
  if(is.na(listing_panel[i,3])& listing_panel[i-1,1]==listing_panel[i,1]){
    listing_panel[i,4] <- xx
  }
  if(is.na(listing_panel[i,3])& listing_panel[i-1,1]!=listing_panel[i,1]){
    xx <- 0
    listing_panel[i,4] <- xx
  }
}
d <- order(listing_panel[,"listing_id"],-listing_panel[,"month"])
listing_panel <- listing_panel[d,]
for(i in 2:nrow(listing_panel)){
  if(listing_panel[i,4]==0){
    listing_panel[i,4]=listing_panel[i-1,4]
  }
}
o <- order(listing_panel[,"listing_id"],listing_panel[,"month"])
listing_panel <- listing_panel[o,]
listing_0price <- listing_panel %>%
  filter(each_month_price == 0)
no_price <- unique(listing_0price$listing_id)

1) Geographic Distribution

Firstly, identifying ‘hotspots’ would be helpful to have an overview of the Airbnb market in Amsterdam. On the one hand, the existing units are usually located in areas where tourists would want to stay, which equates to decent demand. On the other hand, adequate supply means there may be a competition in the market.

Given city government’s concern with too many Airbnb (short-term rental) concentrating in central historic areas, the maps below also help demonstrating how the market indeed is crammed in the central area.

We choose to examine the distribution at two scales – neighborhood (identified by City of Amsterdam) and zip code areas (smaller scale). Both maps show that many Airbnb are located along the canals, which is usually deemed signature of Amsterdam.

sf_use_s2(FALSE)
listings.sf<- listings %>% 
  st_as_sf(coords = c( "longitude","latitude"), crs = 4326, agr = "constant") %>% st_transform('ESRI:102013')

neighbor2 <- neighbor2 %>% 
  st_transform(st_crs(listings.sf))

listing.sf.neighbor2 <- st_intersection(listings.sf,neighbor2) %>% 
  dplyr::select(id, Buurt,Buurt_code) %>% 
  mutate(count=1) %>% 
  st_drop_geometry()

listings.sf <- left_join(listings.sf, listing.sf.neighbor2,by='id')

neighbo2.count <- listings.sf %>% 
  group_by(Buurt) %>%
  dplyr::summarise(airbnb.number = sum(count)) %>% 
  dplyr::select(Buurt, airbnb.number) %>% 
  st_drop_geometry()

neighbor2 <- left_join(neighbor2,neighbo2.count,by="Buurt")

ggplot()+
  geom_sf(data = neighbor2, aes(fill=q5(airbnb.number)),color='transparent')+
  scale_fill_manual(values = Teal,
                     labels = qBr(neighbor2,'airbnb.number'), name = 'Count')+
  labs(title = "Number of Airbnb per Neighborhood", subtitle = "Amsterdam, NL; 2018") +
  mapTheme()

neighbor2 %>% ggplot() + 
      geom_sf(aes(fill = airbnb.number), color = 'white') +
      scale_fill_gradient(low = Teal[1], high = Teal[5],
                          name = "Count") +
      labs(title = "Number of Airbnb per Zipcode", subtitle = "Amsterdam, NL; 2018") +
      mapTheme()

2) Price

Seasonality

The plot below shows the relationship between seasonality and price – we see an annual low in the beginning of the year in Jan and Feb, and a quick incline as spring comes, hitting high points in the summer, than slowly declines as weather gets colder into fall and winter. The price seems to follow the general trend of tourism in Amsterdam.

Month_price <- calendar%>%
  mutate(date2 = ymd(date))%>%
  mutate(month = month(date2),
         listing_id = as.character(listing_id),
         price = parse_number(price))%>%
  mutate(month = as.character(month))%>%
  mutate(month = ifelse(month != 10 & month != 11 & month != 12, gsub("^","0",month), month)) %>%
  drop_na(price) %>%
  group_by(month) %>%
  dplyr::summarize(mean_monthly_price = mean(price))


ggplot(Month_price, 
       aes(x=month, y=mean_monthly_price, group =1)) +
  geom_line(size=1, color = Teal[3]) +
  labs(title = "Average Airbnb Price Every Month", subtitle = "Amsterdam, NL; 2018") +
  xlab("Month of the Year") + ylab("Average Price") +
  plotTheme()

Price per Bed

Since each Airbnb varies in size and number of guests it can accommodate, we choose to visualize price per bed to make a better sense of the price in the market. A bed usually can accommodate 2 guests maximum, though in some case can host up to 3 guests. Again, the higher priced bed are concentrated in the central areas.

details.sf <- 
  details.sf %>% 
  mutate(priceBed=price/beds) %>% 
  filter(priceBed <=1000)


ggplot()+
  geom_sf(data = neighborhood,fill='grey95',color = 'white')+
  geom_sf(data = details.sf, aes(colour=q5(priceBed)),size=.5)+
  scale_color_manual(values = palette5,
                     labels = qBr(details.sf,'priceBed'), name="Price per Bed (Euro)") +
  labs(title = "Price per Bed",
       subtitle = 'Amsterdam Airbnb, price on 2018-12-6') +
  mapTheme()

3) Occupancy

While location and price are helpful information, we shall not disregard an important indicator of actual profit – occupancy. From the existing data, we can see that occupancy is lower in the beginning of the year, then continues to be high year-round. However, it is important to acknowledge the fact that Feburary only has 28 (or 29) days, which may result in lower occupancy count.

index <- function(x, flag = '0') {
  digit <- floor(log10(length(x))) + 1
  paste(flag, formatC(x, width = digit, flag = '0'), sep = '')
}

occupancy <- calendar%>%
  mutate(date2 = ymd(date))%>%
  mutate(month = month(date2),
         count = ifelse(available == "f", 1, 0),
         listing_id = as.character(listing_id))%>%
  filter(!listing_id %in% no_price) %>%
  group_by(listing_id, month) %>%
  dplyr::summarize(monthly_occupancy = sum(count))

monthly_occupancy <- occupancy %>%
  mutate(month = as.character(month))%>%
  mutate(month = ifelse(month != 10 & month != 11 & month != 12, gsub("^","0",month), month)) %>%
  group_by(month) %>%
  dplyr::summarise(mean_monthly_occupancy = mean(monthly_occupancy))

ggplot(monthly_occupancy, 
       aes(x=month, y=mean_monthly_occupancy, group =1)) +
  geom_line(size=1, color = "#FE9900") +
  labs(title = "Airbnb Occupancy Every Month", subtitle = "Amsterdam, NL; 2018") +
  xlab("Month of the Year") + ylab("Average Monthly Occupancy") +
  plotTheme()

4) Composition / Characteristics

Move on to internal features of the units – number of beds, bedrooms, bathrooms, and number of guests it may accommodate – are all fundamentals for determining a unit’s price (and directly associated with guests; willingness to pay). In addition, since Amsterdam has a long history of having boat house and all kinds of special types of houses, tourists may be in favor of experiencing those kinda of accommodation. Hence, we did a thorough examination of all those features and their aggregated effect.

Room Layout

table.basic <- details.sf %>% 
  st_drop_geometry() %>% 
  dplyr::select(price ,beds, bedrooms, bathrooms, accommodates)

stargazer(as.data.frame(table.basic),
          type = "text",
          title ="Table 1. Basic Features Summary",
          single.row = TRUE,
          out.header = TRUE)

## 
## Table 1. Basic Features Summary
## ==================================================
## Statistic      N     Mean   St. Dev.  Min    Max  
## --------------------------------------------------
## price        20,011 150.260 102.648    0    3,900 
## beds         20,011  1.850   1.390     1     32   
## bedrooms     20,004  1.445   0.886     0     12   
## bathrooms    20,001  1.138   0.794   0.000 100.500
## accommodates 20,011  2.892   1.319     1     17   
## --------------------------------------------------

Property Type

count(details.sf$property_type) %>%
  kable() %>%
  kable_styling() %>%
  scroll_box(width = "800px", height = "200px")

x	freq
Aparthotel	72
Apartment	15570
Barn	4
Bed and breakfast	454
Boat	372
Boutique hotel	28
Bungalow	12
Cabin	14
Camper/RV	2
Campsite	2
Casa particular (Cuba)	5
Castle	1
Chalet	3
Condominium	323
Cottage	12
Earth house	1
Guest suite	152
Guesthouse	43
Hostel	4
Hotel	7
House	1520
Houseboat	225
Lighthouse	1
Loft	384
Nature lodge	1
Other	50
Serviced apartment	63
Tent	1
Tiny house	5
Townhouse	648
Villa	32

Puting them together…

After examining individual internal features, we put together the table below to show average price for each combination of all those features. We can see that even with the same number of beds, bathrooms, bedrooms, and no. of accommodating guests, property_type may be a determinant for price.

We have decided to include all those internal features into our machine learning model since they appear to be valid factors.

details.sf %>%
  st_drop_geometry() %>%
  group_by(beds, bathrooms, bedrooms, accommodates, property_type) %>%
  dplyr::summarize(Mean_Price = mean(price)) %>%
  kable() %>%
  kable_styling() %>%
  scroll_box(width = "800px", height = "200px")

## `summarise()` has grouped output by 'beds', 'bathrooms', 'bedrooms',
## 'accommodates'. You can override using the `.groups` argument.

beds	bathrooms	bedrooms	accommodates	property_type	Mean_Price
1	0.0	1	1	Apartment	43.00000
1	0.0	1	1	Townhouse	42.00000
1	0.0	1	2	Apartment	108.50000
1	0.0	1	2	Camper/RV	58.00000
1	0.0	1	2	Guesthouse	125.00000
1	0.0	1	2	House	60.00000
1	0.0	1	3	Boat	50.00000
1	0.0	1	3	Camper/RV	35.00000
1	0.5	0	1	Apartment	73.00000
1	0.5	0	2	Apartment	80.75000
1	0.5	0	2	Bed and breakfast	79.00000
1	0.5	0	2	Boat	50.00000
1	0.5	0	3	Apartment	92.00000
1	0.5	0	4	Guest suite	85.00000
1	0.5	1	1	Apartment	78.66667
1	0.5	1	1	Bed and breakfast	55.00000
1	0.5	1	2	Apartment	91.57143
1	0.5	1	2	Boat	68.00000
1	0.5	1	2	Guest suite	60.00000
1	0.5	1	2	House	72.66667
1	0.5	1	3	Apartment	100.00000
1	1.0	0	1	Apartment	97.94118
1	1.0	0	1	Bed and breakfast	50.00000
1	1.0	0	1	Boutique hotel	79.00000
1	1.0	0	1	House	330.66667
1	1.0	0	2	Aparthotel	133.28571
1	1.0	0	2	Apartment	108.32635
1	1.0	0	2	Bed and breakfast	128.11111
1	1.0	0	2	Boat	110.94118
1	1.0	0	2	Boutique hotel	102.00000
1	1.0	0	2	Bungalow	125.00000
1	1.0	0	2	Cabin	135.00000
1	1.0	0	2	Casa particular (Cuba)	110.00000
1	1.0	0	2	Condominium	104.76923
1	1.0	0	2	Cottage	85.00000
1	1.0	0	2	Guest suite	108.16667
1	1.0	0	2	Guesthouse	117.60000
1	1.0	0	2	House	103.93750
1	1.0	0	2	Houseboat	142.37500
1	1.0	0	2	Loft	124.21212
1	1.0	0	2	Other	95.25000
1	1.0	0	2	Townhouse	112.82353
1	1.0	0	3	Apartment	114.33333
1	1.0	0	3	Bed and breakfast	160.00000
1	1.0	0	3	Boutique hotel	200.00000
1	1.0	0	3	Condominium	140.00000
1	1.0	0	3	Serviced apartment	250.00000
1	1.0	0	4	Apartment	131.61538
1	1.0	0	4	Guest suite	135.00000
1	1.0	0	4	Guesthouse	175.00000
1	1.0	1	1	Apartment	62.96400
1	1.0	1	1	Bed and breakfast	62.77778
1	1.0	1	1	Boat	39.00000
1	1.0	1	1	Boutique hotel	111.66667
1	1.0	1	1	Condominium	51.85714
1	1.0	1	1	Cottage	26.00000
1	1.0	1	1	Guest suite	80.00000
1	1.0	1	1	Guesthouse	103.33333
1	1.0	1	1	Hostel	39.00000
1	1.0	1	1	House	61.88462
1	1.0	1	1	Houseboat	55.00000
1	1.0	1	1	Loft	71.25000
1	1.0	1	1	Other	40.00000
1	1.0	1	1	Serviced apartment	149.00000
1	1.0	1	1	Tiny house	60.00000
1	1.0	1	1	Townhouse	61.70000
1	1.0	1	2	Aparthotel	165.88889
1	1.0	1	2	Apartment	116.43097
1	1.0	1	2	Barn	103.33333
1	1.0	1	2	Bed and breakfast	113.17935
1	1.0	1	2	Boat	137.77647
1	1.0	1	2	Boutique hotel	209.00000
1	1.0	1	2	Bungalow	96.33333
1	1.0	1	2	Cabin	93.20000
1	1.0	1	2	Campsite	70.00000
1	1.0	1	2	Casa particular (Cuba)	85.00000
1	1.0	1	2	Castle	150.00000
1	1.0	1	2	Chalet	83.00000
1	1.0	1	2	Condominium	114.64348
1	1.0	1	2	Cottage	128.25000
1	1.0	1	2	Earth house	80.00000
1	1.0	1	2	Guest suite	105.93750
1	1.0	1	2	Guesthouse	114.43750
1	1.0	1	2	Hotel	435.00000
1	1.0	1	2	House	94.42069
1	1.0	1	2	Houseboat	140.42373
1	1.0	1	2	Lighthouse	850.00000
1	1.0	1	2	Loft	151.63115
1	1.0	1	2	Other	93.96000
1	1.0	1	2	Serviced apartment	186.40909
1	1.0	1	2	Tent	75.00000
1	1.0	1	2	Tiny house	76.33333
1	1.0	1	2	Townhouse	95.71795
1	1.0	1	2	Villa	88.00000
1	1.0	1	3	Apartment	122.08835
1	1.0	1	3	Bed and breakfast	128.40000
1	1.0	1	3	Boat	176.66667
1	1.0	1	3	Condominium	80.00000
1	1.0	1	3	House	127.50000
1	1.0	1	3	Houseboat	151.80000
1	1.0	1	3	Loft	180.33333
1	1.0	1	3	Townhouse	60.25000
1	1.0	1	4	Aparthotel	164.00000
1	1.0	1	4	Apartment	147.65000
1	1.0	1	4	Bed and breakfast	299.00000
1	1.0	1	4	Boat	165.00000
1	1.0	1	4	Condominium	149.00000
1	1.0	1	4	Guest suite	165.00000
1	1.0	1	4	Guesthouse	250.00000
1	1.0	1	4	Hotel	69.00000
1	1.0	1	4	House	420.00000
1	1.0	1	4	Houseboat	325.00000
1	1.0	1	4	Loft	150.00000
1	1.0	1	4	Townhouse	138.33333
1	1.0	1	5	Apartment	104.50000
1	1.0	1	6	Apartment	130.00000
1	1.0	1	6	House	325.00000
1	1.0	2	1	Apartment	27.00000
1	1.0	2	1	Bed and breakfast	121.00000
1	1.0	2	2	Apartment	126.78102
1	1.0	2	2	Condominium	123.40000
1	1.0	2	2	House	106.63636
1	1.0	2	2	Houseboat	105.00000
1	1.0	2	2	Loft	180.00000
1	1.0	2	3	Apartment	136.09375
1	1.0	2	3	Condominium	100.00000
1	1.0	2	3	House	99.00000
1	1.0	2	3	Townhouse	100.00000
1	1.0	2	4	Apartment	167.57143
1	1.0	2	4	House	112.66667
1	1.0	2	4	Townhouse	165.00000
1	1.0	2	5	Apartment	205.50000
1	1.0	3	2	Apartment	91.00000
1	1.0	3	2	Other	295.00000
1	1.0	3	3	Apartment	212.25000
1	1.0	3	4	Apartment	142.50000
1	1.0	3	4	Loft	150.00000
1	1.0	3	4	Townhouse	130.00000
1	1.0	3	5	Townhouse	185.00000
1	1.0	4	1	Apartment	45.00000
1	1.0	4	2	Apartment	94.00000
1	1.0	NA	2	Apartment	123.00000
1	1.5	0	2	Apartment	118.85714
1	1.5	0	2	Boat	155.00000
1	1.5	0	2	Loft	200.00000
1	1.5	0	3	Apartment	85.00000
1	1.5	1	1	Apartment	65.04348
1	1.5	1	1	Bed and breakfast	77.85714
1	1.5	1	1	Condominium	90.00000
1	1.5	1	1	House	45.60000
1	1.5	1	2	Aparthotel	152.60000
1	1.5	1	2	Apartment	122.01389
1	1.5	1	2	Bed and breakfast	93.85714
1	1.5	1	2	Boat	184.00000
1	1.5	1	2	Cabin	80.00000
1	1.5	1	2	Condominium	116.00000
1	1.5	1	2	Guest suite	88.60000
1	1.5	1	2	House	86.02857
1	1.5	1	2	Houseboat	80.00000
1	1.5	1	2	Loft	131.81250
1	1.5	1	2	Other	245.00000
1	1.5	1	2	Serviced apartment	75.00000
1	1.5	1	2	Townhouse	100.10526
1	1.5	1	3	Apartment	140.06667
1	1.5	1	3	House	100.00000
1	1.5	1	4	Aparthotel	190.00000
1	1.5	1	4	Apartment	171.40000
1	1.5	1	4	Bed and breakfast	399.00000
1	1.5	1	4	Boat	420.00000
1	1.5	2	1	Apartment	129.00000
1	1.5	2	2	Apartment	129.09524
1	1.5	2	2	House	457.00000
1	1.5	2	2	Townhouse	180.00000
1	1.5	2	3	Apartment	166.00000
1	1.5	2	4	Apartment	162.62500
1	1.5	2	4	Townhouse	100.00000
1	1.5	2	5	Apartment	90.00000
1	1.5	2	8	Apartment	80.00000
1	1.5	3	4	Apartment	250.00000
1	2.0	0	1	Apartment	49.00000
1	2.0	1	1	Apartment	60.66667
1	2.0	1	1	House	40.00000
1	2.0	1	1	Townhouse	60.00000
1	2.0	1	2	Apartment	104.51852
1	2.0	1	2	Bed and breakfast	70.00000
1	2.0	1	2	Hostel	85.00000
1	2.0	1	2	House	62.50000
1	2.0	1	2	Serviced apartment	95.00000
1	2.0	1	2	Townhouse	91.80000
1	2.0	1	3	Apartment	225.00000
1	2.0	1	3	House	100.00000
1	2.0	1	4	Bed and breakfast	279.00000
1	2.0	2	2	Apartment	165.00000
1	2.0	2	4	Apartment	650.00000
1	2.5	1	1	Apartment	65.00000
1	2.5	1	2	Apartment	93.80000
1	2.5	1	2	House	138.33333
1	2.5	2	2	Apartment	180.00000
1	3.0	1	1	Apartment	80.00000
1	3.0	1	2	Condominium	75.00000
1	3.0	1	2	House	60.00000
1	3.0	3	4	House	300.00000
1	4.0	1	1	Apartment	50.00000
1	7.0	1	2	Apartment	95.00000
1	NA	0	1	House	900.00000
1	NA	1	2	Apartment	89.75000
1	NA	1	2	Bed and breakfast	90.00000
1	NA	1	4	Apartment	175.00000
1	NA	2	4	Apartment	750.00000
2	0.0	1	2	Apartment	71.00000
2	0.0	1	2	Boat	75.00000
2	0.5	0	4	Condominium	169.00000
2	0.5	1	2	Apartment	87.00000
2	0.5	1	2	Bed and breakfast	95.00000
2	0.5	1	2	Boat	50.00000
2	0.5	1	2	Boutique hotel	120.00000
2	0.5	1	3	Apartment	130.00000
2	0.5	1	3	Boat	80.00000
2	0.5	1	3	House	60.00000
2	0.5	1	4	Apartment	100.75000
2	0.5	1	4	Guest suite	85.00000
2	0.5	2	4	Apartment	185.00000
2	0.5	2	4	House	185.00000
2	1.0	0	2	Apartment	109.51923
2	1.0	0	2	Bed and breakfast	110.00000
2	1.0	0	2	Boat	120.00000
2	1.0	0	2	Cabin	75.00000
2	1.0	0	2	Condominium	169.00000
2	1.0	0	2	Guest suite	68.66667
2	1.0	0	2	Guesthouse	75.00000
2	1.0	0	2	House	108.33333
2	1.0	0	2	Loft	125.00000
2	1.0	0	2	Other	95.00000
2	1.0	0	2	Serviced apartment	249.00000
2	1.0	0	2	Townhouse	90.00000
2	1.0	0	3	Apartment	123.69565
2	1.0	0	3	Bed and breakfast	116.75000
2	1.0	0	3	Boat	75.00000
2	1.0	0	3	Cabin	120.00000
2	1.0	0	3	Guest suite	125.00000
2	1.0	0	3	Guesthouse	80.00000
2	1.0	0	3	Houseboat	97.50000
2	1.0	0	3	Loft	230.00000
2	1.0	0	3	Townhouse	94.00000
2	1.0	0	4	Apartment	151.66667
2	1.0	0	4	Bed and breakfast	170.00000
2	1.0	0	4	Boat	172.00000
2	1.0	0	4	Cabin	149.00000
2	1.0	0	4	Condominium	90.00000
2	1.0	0	4	Guest suite	95.00000
2	1.0	0	4	House	204.25000
2	1.0	0	4	Houseboat	85.00000
2	1.0	0	4	Loft	135.00000
2	1.0	0	4	Other	86.00000
2	1.0	0	4	Townhouse	110.00000
2	1.0	0	5	Apartment	130.00000
2	1.0	1	1	Apartment	77.80000
2	1.0	1	1	House	57.50000
2	1.0	1	2	Aparthotel	172.00000
2	1.0	1	2	Apartment	119.33259
2	1.0	1	2	Bed and breakfast	99.82979
2	1.0	1	2	Boat	100.15385
2	1.0	1	2	Boutique hotel	91.20000
2	1.0	1	2	Condominium	90.57143
2	1.0	1	2	Guest suite	108.40000
2	1.0	1	2	Guesthouse	89.50000
2	1.0	1	2	Hotel	140.00000
2	1.0	1	2	House	94.09375
2	1.0	1	2	Houseboat	181.57143
2	1.0	1	2	Loft	151.14286
2	1.0	1	2	Other	104.00000
2	1.0	1	2	Serviced apartment	84.00000
2	1.0	1	2	Tiny house	110.00000
2	1.0	1	2	Townhouse	99.90000
2	1.0	1	3	Apartment	125.95077
2	1.0	1	3	Bed and breakfast	138.00000
2	1.0	1	3	Boat	108.00000
2	1.0	1	3	Bungalow	95.00000
2	1.0	1	3	Cabin	85.00000
2	1.0	1	3	Condominium	178.83333
2	1.0	1	3	Guest suite	129.66667
2	1.0	1	3	House	108.41176
2	1.0	1	3	Houseboat	187.00000
2	1.0	1	3	Loft	160.50000
2	1.0	1	3	Nature lodge	140.00000
2	1.0	1	3	Serviced apartment	179.00000
2	1.0	1	3	Townhouse	123.11111
2	1.0	1	3	Villa	75.00000
2	1.0	1	4	Aparthotel	167.00000
2	1.0	1	4	Apartment	152.66940
2	1.0	1	4	Bed and breakfast	176.05882
2	1.0	1	4	Boat	204.46154
2	1.0	1	4	Boutique hotel	180.00000
2	1.0	1	4	Condominium	133.44444
2	1.0	1	4	Guest suite	133.25000
2	1.0	1	4	Guesthouse	105.00000
2	1.0	1	4	House	141.18182
2	1.0	1	4	Houseboat	187.70000
2	1.0	1	4	Loft	195.11111
2	1.0	1	4	Other	220.00000
2	1.0	1	4	Serviced apartment	150.00000
2	1.0	1	4	Townhouse	107.50000
2	1.0	2	1	Apartment	92.50000
2	1.0	2	2	Apartment	132.28082
2	1.0	2	2	Bed and breakfast	72.50000
2	1.0	2	2	Boat	188.00000
2	1.0	2	2	Condominium	113.00000
2	1.0	2	2	House	119.44444
2	1.0	2	2	Houseboat	345.00000
2	1.0	2	2	Loft	250.00000
2	1.0	2	2	Other	220.00000
2	1.0	2	2	Townhouse	150.00000
2	1.0	2	3	Apartment	145.11750
2	1.0	2	3	Bed and breakfast	112.00000
2	1.0	2	3	Boat	161.62500
2	1.0	2	3	Condominium	145.25000
2	1.0	2	3	Guest suite	149.00000
2	1.0	2	3	Guesthouse	99.00000
2	1.0	2	3	House	152.65000
2	1.0	2	3	Houseboat	215.00000
2	1.0	2	3	Loft	198.20000
2	1.0	2	3	Other	142.33333
2	1.0	2	3	Serviced apartment	160.00000
2	1.0	2	3	Townhouse	123.11111
2	1.0	2	4	Aparthotel	204.00000
2	1.0	2	4	Apartment	181.22114
2	1.0	2	4	Bed and breakfast	191.50000
2	1.0	2	4	Boat	204.41667
2	1.0	2	4	Condominium	187.29167
2	1.0	2	4	Guest suite	217.00000
2	1.0	2	4	House	172.63725
2	1.0	2	4	Houseboat	178.57895
2	1.0	2	4	Loft	259.46667
2	1.0	2	4	Serviced apartment	215.00000
2	1.0	2	4	Townhouse	215.37838
2	1.0	2	5	Apartment	186.76000
2	1.0	2	5	Boat	300.00000
2	1.0	2	5	House	130.00000
2	1.0	2	6	Apartment	220.62500
2	1.0	3	2	Apartment	109.50000
2	1.0	3	2	Townhouse	210.00000
2	1.0	3	3	Apartment	127.50000
2	1.0	3	3	House	150.00000
2	1.0	3	3	Townhouse	120.00000
2	1.0	3	4	Apartment	174.42308
2	1.0	3	4	Condominium	100.00000
2	1.0	3	4	House	149.42857
2	1.0	3	4	Houseboat	150.00000
2	1.0	3	4	Loft	800.00000
2	1.0	3	4	Townhouse	250.00000
2	1.0	3	5	Apartment	163.33333
2	1.0	3	5	House	259.00000
2	1.0	4	4	House	145.00000
2	1.5	0	2	Bed and breakfast	107.50000
2	1.5	0	3	Apartment	95.00000
2	1.5	0	3	Guest suite	190.00000
2	1.5	0	3	Guesthouse	115.00000
2	1.5	0	3	Loft	179.00000
2	1.5	0	4	Bed and breakfast	80.00000
2	1.5	1	1	Apartment	45.00000
2	1.5	1	1	Condominium	90.00000
2	1.5	1	2	Aparthotel	167.00000
2	1.5	1	2	Apartment	127.38596
2	1.5	1	2	Bed and breakfast	102.40000
2	1.5	1	2	Boat	239.50000
2	1.5	1	2	Guest suite	83.50000
2	1.5	1	2	Hotel	120.00000
2	1.5	1	2	House	95.20000
2	1.5	1	2	Other	90.00000
2	1.5	1	2	Serviced apartment	109.00000
2	1.5	1	2	Townhouse	98.50000
2	1.5	1	3	Apartment	141.38462
2	1.5	1	3	Condominium	108.00000
2	1.5	1	3	House	90.00000
2	1.5	1	3	Houseboat	99.00000
2	1.5	1	3	Loft	142.50000
2	1.5	1	3	Townhouse	107.50000
2	1.5	1	4	Aparthotel	225.00000
2	1.5	1	4	Apartment	152.95000
2	1.5	1	4	Bed and breakfast	240.00000
2	1.5	1	4	Condominium	150.00000
2	1.5	1	4	House	108.00000
2	1.5	1	4	Houseboat	150.00000
2	1.5	1	4	Serviced apartment	179.00000
2	1.5	1	6	Apartment	181.00000
2	1.5	2	2	Apartment	135.12500
2	1.5	2	2	House	137.50000
2	1.5	2	2	Townhouse	150.00000
2	1.5	2	3	Apartment	157.16981
2	1.5	2	3	House	110.71429
2	1.5	2	3	Houseboat	180.00000
2	1.5	2	3	Loft	225.00000
2	1.5	2	3	Townhouse	137.50000
2	1.5	2	4	Aparthotel	265.00000
2	1.5	2	4	Apartment	199.80531
2	1.5	2	4	Bed and breakfast	490.00000
2	1.5	2	4	Boat	185.00000
2	1.5	2	4	Boutique hotel	190.00000
2	1.5	2	4	Casa particular (Cuba)	180.00000
2	1.5	2	4	Condominium	169.75000
2	1.5	2	4	Guest suite	149.50000
2	1.5	2	4	Guesthouse	104.50000
2	1.5	2	4	House	202.88889
2	1.5	2	4	Houseboat	200.80000
2	1.5	2	4	Loft	310.36364
2	1.5	2	4	Serviced apartment	212.00000
2	1.5	2	4	Townhouse	210.20000
2	1.5	2	5	Apartment	245.16667
2	1.5	2	5	House	150.00000
2	1.5	2	6	Aparthotel	225.00000
2	1.5	2	6	Apartment	276.00000
2	1.5	3	3	Apartment	143.33333
2	1.5	3	4	Apartment	218.00000
2	1.5	3	4	Condominium	104.50000
2	1.5	3	4	House	113.33333
2	1.5	3	6	Apartment	295.00000
2	1.5	4	4	Apartment	190.00000
2	2.0	0	4	Apartment	200.00000
2	2.0	0	4	Townhouse	190.00000
2	2.0	1	2	Apartment	97.00000
2	2.0	1	2	Bed and breakfast	120.00000
2	2.0	1	2	House	60.00000
2	2.0	1	3	Apartment	100.00000
2	2.0	1	3	House	63.00000
2	2.0	1	4	Aparthotel	149.00000
2	2.0	1	4	Apartment	164.75000
2	2.0	1	4	House	154.50000
2	2.0	1	4	Other	159.00000
2	2.0	1	4	Townhouse	222.00000
2	2.0	2	2	Apartment	182.00000
2	2.0	2	2	Townhouse	190.00000
2	2.0	2	3	Apartment	137.16667
2	2.0	2	3	Houseboat	250.00000
2	2.0	2	3	Loft	180.00000
2	2.0	2	4	Apartment	247.03906
2	2.0	2	4	Bed and breakfast	227.57143
2	2.0	2	4	Boat	249.80000
2	2.0	2	4	Casa particular (Cuba)	400.00000
2	2.0	2	4	Condominium	257.25000
2	2.0	2	4	Guest suite	238.25000
2	2.0	2	4	House	197.61905
2	2.0	2	4	Houseboat	325.00000
2	2.0	2	4	Loft	279.00000
2	2.0	2	4	Serviced apartment	226.00000
2	2.0	2	4	Townhouse	256.50000
2	2.0	2	4	Villa	149.00000
2	2.0	2	5	Apartment	140.00000
2	2.0	2	6	House	190.00000
2	2.0	3	4	Apartment	281.25000
2	2.0	3	4	House	256.60000
2	2.0	3	6	Townhouse	189.50000
2	2.5	1	3	Apartment	150.00000
2	2.5	2	2	Apartment	150.00000
2	2.5	2	3	Apartment	200.00000
2	2.5	2	4	Apartment	251.55000
2	2.5	2	4	Guest suite	345.00000
2	2.5	2	4	House	195.00000
2	2.5	2	4	Houseboat	225.00000
2	2.5	2	4	Loft	212.00000
2	2.5	2	4	Serviced apartment	590.00000
2	2.5	2	4	Townhouse	512.50000
2	2.5	3	4	Apartment	245.00000
2	2.5	3	4	House	61.00000
2	3.0	1	2	Apartment	187.00000
2	3.0	1	2	Boat	51.42857
2	3.0	2	2	Apartment	70.00000
2	3.0	2	4	Apartment	400.00000
2	3.0	2	4	Condominium	350.00000
2	3.0	3	4	Apartment	175.00000
2	3.0	3	4	Townhouse	220.00000
2	3.5	2	4	Apartment	422.50000
2	3.5	2	4	House	90.00000
2	4.0	1	4	Apartment	60.00000
2	8.0	0	2	Condominium	120.00000
2	15.0	1	4	Apartment	175.00000
3	0.0	0	4	Boat	163.00000
3	0.0	1	4	Loft	107.00000
3	0.5	0	3	Boat	60.00000
3	0.5	0	4	Guest suite	125.00000
3	0.5	1	2	Boat	35.00000
3	0.5	1	3	Bed and breakfast	140.00000
3	0.5	1	3	Boat	59.00000
3	0.5	1	4	Bed and breakfast	54.00000
3	0.5	1	4	Boat	150.00000
3	0.5	3	4	Apartment	270.00000
3	1.0	0	2	Apartment	144.33333
3	1.0	0	3	Apartment	86.50000
3	1.0	0	3	Bed and breakfast	75.00000
3	1.0	0	3	Boat	195.00000
3	1.0	0	3	Chalet	80.00000
3	1.0	0	3	Guest suite	77.50000
3	1.0	0	3	Loft	99.00000
3	1.0	0	4	Aparthotel	210.00000
3	1.0	0	4	Apartment	140.66667
3	1.0	0	4	Bed and breakfast	120.00000
3	1.0	0	4	Cabin	160.00000
3	1.0	0	4	Cottage	115.00000
3	1.0	0	4	Guest suite	75.00000
3	1.0	0	4	House	199.00000
3	1.0	0	4	Houseboat	150.00000
3	1.0	0	4	Loft	240.00000
3	1.0	0	5	Apartment	85.00000
3	1.0	0	6	Apartment	190.00000
3	1.0	1	2	Apartment	130.55556
3	1.0	1	2	Boat	95.00000
3	1.0	1	2	Houseboat	199.00000
3	1.0	1	2	Loft	70.00000
3	1.0	1	2	Townhouse	159.66667
3	1.0	1	3	Apartment	129.62222
3	1.0	1	3	Bed and breakfast	96.00000
3	1.0	1	3	Boat	117.00000
3	1.0	1	3	Boutique hotel	79.00000
3	1.0	1	3	Condominium	110.00000
3	1.0	1	3	Guest suite	80.00000
3	1.0	1	3	Hotel	165.00000
3	1.0	1	3	House	135.00000
3	1.0	1	3	Loft	84.50000
3	1.0	1	3	Townhouse	85.00000
3	1.0	1	4	Aparthotel	203.25000
3	1.0	1	4	Apartment	168.18462
3	1.0	1	4	Bed and breakfast	139.66667
3	1.0	1	4	Boat	175.00000
3	1.0	1	4	Condominium	154.00000
3	1.0	1	4	Guest suite	60.00000
3	1.0	1	4	House	106.33333
3	1.0	1	4	Houseboat	462.50000
3	1.0	1	4	Loft	307.50000
3	1.0	1	4	Serviced apartment	149.00000
3	1.0	1	4	Townhouse	120.00000
3	1.0	1	5	Apartment	152.00000
3	1.0	1	5	Bed and breakfast	150.00000
3	1.0	1	5	House	161.33333
3	1.0	1	6	Apartment	415.00000
3	1.0	1	12	Apartment	262.50000
3	1.0	2	2	Apartment	136.41667
3	1.0	2	2	Boat	70.00000
3	1.0	2	2	Guesthouse	105.00000
3	1.0	2	2	House	140.00000
3	1.0	2	2	Houseboat	195.00000
3	1.0	2	2	Loft	162.50000
3	1.0	2	3	Apartment	148.43939
3	1.0	2	3	Boat	191.66667
3	1.0	2	3	House	143.16667
3	1.0	2	3	Houseboat	250.00000
3	1.0	2	3	Loft	179.00000
3	1.0	2	3	Other	135.00000
3	1.0	2	3	Townhouse	117.00000
3	1.0	2	4	Aparthotel	239.37500
3	1.0	2	4	Apartment	198.21034
3	1.0	2	4	Bed and breakfast	136.00000
3	1.0	2	4	Boat	228.78571
3	1.0	2	4	Bungalow	140.00000
3	1.0	2	4	Condominium	205.00000
3	1.0	2	4	Guest suite	85.00000
3	1.0	2	4	House	147.90323
3	1.0	2	4	Houseboat	238.30769
3	1.0	2	4	Loft	226.60000
3	1.0	2	4	Serviced apartment	349.50000
3	1.0	2	4	Townhouse	175.78947
3	1.0	2	4	Villa	200.00000
3	1.0	2	5	Apartment	200.33333
3	1.0	2	5	House	182.00000
3	1.0	2	5	Houseboat	225.00000
3	1.0	2	6	Apartment	238.44444
3	1.0	2	6	Loft	550.00000
3	1.0	2	7	Apartment	191.00000
3	1.0	3	1	Apartment	40.00000
3	1.0	3	2	Apartment	143.22222
3	1.0	3	2	House	123.33333
3	1.0	3	3	Apartment	161.18182
3	1.0	3	3	House	151.00000
3	1.0	3	3	Townhouse	120.00000
3	1.0	3	4	Apartment	204.22581
3	1.0	3	4	Boat	252.30000
3	1.0	3	4	Condominium	205.00000
3	1.0	3	4	Cottage	95.00000
3	1.0	3	4	House	179.15789
3	1.0	3	4	Houseboat	263.33333
3	1.0	3	4	Loft	311.85714
3	1.0	3	4	Townhouse	190.62500
3	1.0	3	4	Villa	225.00000
3	1.0	3	5	Apartment	194.50000
3	1.0	3	5	House	183.52632
3	1.0	3	5	Townhouse	122.33333
3	1.0	3	6	Apartment	260.14706
3	1.0	3	6	Boat	394.00000
3	1.0	3	6	Condominium	390.00000
3	1.0	3	6	House	295.50000
3	1.0	3	6	Loft	450.00000
3	1.0	3	6	Other	225.00000
3	1.0	3	6	Townhouse	160.00000
3	1.0	3	7	Apartment	248.00000
3	1.0	3	7	Townhouse	224.00000
3	1.0	3	8	Apartment	450.00000
3	1.0	3	8	House	420.00000
3	1.0	4	4	Apartment	209.66667
3	1.0	4	4	House	131.00000
3	1.0	4	5	Apartment	131.66667
3	1.0	NA	4	Apartment	179.00000
3	1.5	0	6	Boat	98.00000
3	1.5	1	3	Apartment	12.00000
3	1.5	1	3	Bed and breakfast	180.00000
3	1.5	1	4	Apartment	158.13333
3	1.5	1	4	Houseboat	176.00000
3	1.5	1	5	Apartment	90.00000
3	1.5	2	2	Apartment	149.33333
3	1.5	2	2	Houseboat	175.00000
3	1.5	2	3	Apartment	138.00000
3	1.5	2	3	Boat	190.00000
3	1.5	2	3	Condominium	90.00000
3	1.5	2	3	House	149.50000
3	1.5	2	3	Townhouse	354.00000
3	1.5	2	4	Aparthotel	224.85714
3	1.5	2	4	Apartment	209.12791
3	1.5	2	4	Boat	77.50000
3	1.5	2	4	Condominium	365.00000
3	1.5	2	4	House	164.42857
3	1.5	2	4	Houseboat	453.00000
3	1.5	2	4	Loft	209.00000
3	1.5	2	4	Serviced apartment	500.00000
3	1.5	2	4	Townhouse	192.80000
3	1.5	2	5	Apartment	244.14286
3	1.5	2	5	Guest suite	250.00000
3	1.5	2	6	Apartment	252.33333
3	1.5	2	6	Condominium	191.00000
3	1.5	2	6	House	169.00000
3	1.5	3	2	Apartment	134.66667
3	1.5	3	3	Apartment	192.00000
3	1.5	3	3	House	130.00000
3	1.5	3	4	Apartment	226.21538
3	1.5	3	4	Boat	157.50000
3	1.5	3	4	House	193.22222
3	1.5	3	4	Houseboat	325.00000
3	1.5	3	4	Loft	300.00000
3	1.5	3	4	Townhouse	211.40000
3	1.5	3	5	Apartment	275.81818
3	1.5	3	5	House	296.42857
3	1.5	3	5	Townhouse	220.00000
3	1.5	3	6	Apartment	195.90000
3	1.5	3	6	House	269.50000
3	1.5	3	6	Houseboat	249.00000
3	1.5	3	6	Townhouse	347.50000
3	1.5	3	7	Apartment	298.00000
3	1.5	3	7	Houseboat	280.00000
3	1.5	3	8	Apartment	650.00000
3	1.5	4	3	Apartment	139.00000
3	1.5	4	4	Apartment	200.00000
3	1.5	4	4	House	200.00000
3	1.5	4	4	Townhouse	125.00000
3	1.5	4	6	Apartment	244.00000
3	2.0	0	4	Apartment	125.00000
3	2.0	1	3	Loft	190.00000
3	2.0	1	4	Apartment	95.00000
3	2.0	1	4	Guest suite	145.00000
3	2.0	1	4	Serviced apartment	149.00000
3	2.0	2	3	Apartment	195.33333
3	2.0	2	3	Bed and breakfast	125.00000
3	2.0	2	4	Apartment	252.00000
3	2.0	2	4	Bed and breakfast	205.00000
3	2.0	2	4	Boat	188.50000
3	2.0	2	4	Condominium	250.00000
3	2.0	2	4	Guest suite	195.00000
3	2.0	2	4	House	226.25000
3	2.0	2	4	Houseboat	275.00000
3	2.0	2	4	Loft	320.00000
3	2.0	2	4	Townhouse	291.00000
3	2.0	2	5	Apartment	302.50000
3	2.0	2	5	House	295.00000
3	2.0	2	5	Houseboat	335.00000
3	2.0	2	5	Townhouse	155.00000
3	2.0	2	6	Apartment	350.00000
3	2.0	2	6	House	355.00000
3	2.0	3	3	House	74.00000
3	2.0	3	4	Apartment	263.32432
3	2.0	3	4	Boat	157.00000
3	2.0	3	4	Condominium	190.00000
3	2.0	3	4	House	220.61538
3	2.0	3	4	Houseboat	638.00000
3	2.0	3	4	Loft	499.66667
3	2.0	3	4	Serviced apartment	285.00000
3	2.0	3	4	Townhouse	325.23077
3	2.0	3	5	Apartment	258.00000
3	2.0	3	5	House	184.50000
3	2.0	3	6	Apartment	359.15385
3	2.0	3	6	Boat	299.00000
3	2.0	3	6	House	199.00000
3	2.0	3	6	Loft	575.00000
3	2.0	3	6	Townhouse	270.00000
3	2.0	3	7	Townhouse	150.00000
3	2.0	3	8	Apartment	450.00000
3	2.0	4	4	Apartment	149.00000
3	2.0	4	4	Houseboat	200.00000
3	2.0	4	6	Apartment	450.00000
3	2.5	2	4	Apartment	221.50000
3	2.5	2	4	Bed and breakfast	260.00000
3	2.5	2	4	House	270.00000
3	2.5	2	5	Loft	400.00000
3	2.5	3	4	Apartment	200.37500
3	2.5	3	4	Bed and breakfast	380.00000
3	2.5	3	4	House	218.00000
3	2.5	3	4	Loft	198.00000
3	2.5	3	4	Townhouse	277.20000
3	2.5	3	5	Apartment	250.00000
3	2.5	3	5	House	600.00000
3	2.5	3	6	Apartment	407.50000
3	2.5	3	6	House	274.50000
3	2.5	3	6	Townhouse	379.50000
3	2.5	3	7	Apartment	400.00000
3	2.5	4	4	Apartment	212.00000
3	2.5	4	4	Townhouse	475.00000
3	3.0	3	4	Apartment	375.00000
3	3.0	3	4	House	225.00000
3	3.0	3	4	Townhouse	360.00000
3	3.0	3	6	Apartment	600.00000
3	3.0	3	8	Apartment	121.00000
3	3.0	4	5	Apartment	375.00000
3	3.5	1	3	House	65.00000
3	3.5	3	4	House	350.00000
3	3.5	3	6	Loft	350.00000
4	0.0	2	4	Boat	175.00000
4	0.5	1	3	Apartment	125.00000
4	0.5	1	4	Bed and breakfast	109.00000
4	0.5	2	4	Boat	291.00000
4	1.0	0	4	Apartment	160.50000
4	1.0	0	4	Bed and breakfast	195.00000
4	1.0	0	4	Boat	93.00000
4	1.0	0	4	Boutique hotel	190.00000
4	1.0	0	4	Houseboat	200.00000
4	1.0	0	4	Loft	185.00000
4	1.0	0	4	Serviced apartment	299.00000
4	1.0	1	2	Apartment	133.00000
4	1.0	1	3	Apartment	475.00000
4	1.0	1	3	Bed and breakfast	160.00000
4	1.0	1	4	Aparthotel	302.00000
4	1.0	1	4	Apartment	188.06250
4	1.0	1	4	Bed and breakfast	138.00000
4	1.0	1	4	Boat	166.50000
4	1.0	1	4	Cabin	75.00000
4	1.0	1	4	Condominium	115.00000
4	1.0	1	4	Guesthouse	200.00000
4	1.0	1	4	House	131.87500
4	1.0	1	4	Houseboat	259.00000
4	1.0	1	4	Loft	360.00000
4	1.0	1	4	Townhouse	106.00000
4	1.0	1	6	Apartment	287.50000
4	1.0	1	8	Apartment	600.00000
4	1.0	2	2	Apartment	115.66667
4	1.0	2	2	House	135.00000
4	1.0	2	3	Apartment	124.75000
4	1.0	2	3	House	100.00000
4	1.0	2	4	Aparthotel	160.00000
4	1.0	2	4	Apartment	199.28750
4	1.0	2	4	Bed and breakfast	207.00000
4	1.0	2	4	Boat	226.58824
4	1.0	2	4	Cabin	200.00000
4	1.0	2	4	Chalet	90.00000
4	1.0	2	4	Condominium	133.33333
4	1.0	2	4	Cottage	125.00000
4	1.0	2	4	Guest suite	145.00000
4	1.0	2	4	Guesthouse	130.00000
4	1.0	2	4	House	181.69444
4	1.0	2	4	Houseboat	211.33333
4	1.0	2	4	Loft	234.28571
4	1.0	2	4	Serviced apartment	260.00000
4	1.0	2	4	Townhouse	191.08333
4	1.0	2	4	Villa	300.00000
4	1.0	2	5	Apartment	171.66667
4	1.0	2	5	Boat	250.00000
4	1.0	2	5	Condominium	250.00000
4	1.0	2	5	House	200.00000
4	1.0	2	5	Houseboat	150.00000
4	1.0	2	6	Aparthotel	269.00000
4	1.0	2	6	Apartment	470.00000
4	1.0	2	6	Townhouse	350.00000
4	1.0	3	2	Apartment	139.00000
4	1.0	3	3	Houseboat	180.00000
4	1.0	3	4	Aparthotel	244.00000
4	1.0	3	4	Apartment	197.32500
4	1.0	3	4	Bed and breakfast	320.00000
4	1.0	3	4	Boat	167.83333
4	1.0	3	4	Condominium	249.66667
4	1.0	3	4	Cottage	225.00000
4	1.0	3	4	House	167.95000
4	1.0	3	4	Houseboat	183.40000
4	1.0	3	4	Loft	181.66667
4	1.0	3	4	Serviced apartment	150.00000
4	1.0	3	4	Townhouse	186.55000
4	1.0	3	5	Apartment	257.58333
4	1.0	3	5	Boat	200.00000
4	1.0	3	5	Condominium	169.00000
4	1.0	3	5	House	176.80000
4	1.0	3	5	Loft	175.00000
4	1.0	3	5	Townhouse	117.00000
4	1.0	3	6	Apartment	248.77778
4	1.0	3	6	Condominium	127.50000
4	1.0	3	6	House	250.00000
4	1.0	3	6	Loft	295.00000
4	1.0	3	6	Townhouse	198.50000
4	1.0	3	7	Apartment	325.00000
4	1.0	3	7	House	215.00000
4	1.0	3	8	Apartment	236.00000
4	1.0	4	4	Apartment	221.72222
4	1.0	4	4	Condominium	350.00000
4	1.0	4	4	House	204.12500
4	1.0	4	4	Loft	275.00000
4	1.0	4	4	Townhouse	202.37500
4	1.0	4	5	Apartment	137.50000
4	1.0	4	5	House	138.75000
4	1.0	4	5	Townhouse	237.00000
4	1.0	4	6	Apartment	212.66667
4	1.0	4	6	House	228.40000
4	1.0	4	6	Houseboat	112.50000
4	1.0	4	6	Townhouse	256.66667
4	1.0	4	7	House	256.00000
4	1.0	4	7	Townhouse	400.00000
4	1.0	4	8	Apartment	400.00000
4	1.0	4	8	House	250.00000
4	1.5	1	4	Apartment	162.90000
4	1.5	1	4	Boat	117.50000
4	1.5	1	4	Loft	217.50000
4	1.5	2	2	Apartment	100.00000
4	1.5	2	2	Serviced apartment	299.00000
4	1.5	2	3	Apartment	100.00000
4	1.5	2	3	Condominium	74.00000
4	1.5	2	3	House	135.00000
4	1.5	2	4	Apartment	259.79104
4	1.5	2	4	Boat	250.00000
4	1.5	2	4	Boutique hotel	190.00000
4	1.5	2	4	Condominium	120.50000
4	1.5	2	4	Cottage	143.00000
4	1.5	2	4	Guest suite	127.00000
4	1.5	2	4	House	226.40000
4	1.5	2	4	Houseboat	299.00000
4	1.5	2	4	Loft	250.00000
4	1.5	2	4	Townhouse	245.80000
4	1.5	2	4	Villa	175.00000
4	1.5	2	5	Apartment	195.00000
4	1.5	2	5	House	125.00000
4	1.5	2	5	Townhouse	115.00000
4	1.5	2	6	Apartment	382.50000
4	1.5	2	6	House	230.00000
4	1.5	2	6	Serviced apartment	265.00000
4	1.5	2	7	Apartment	363.66667
4	1.5	3	2	Apartment	109.00000
4	1.5	3	3	Apartment	139.00000
4	1.5	3	4	Apartment	187.75610
4	1.5	3	4	Boat	175.00000
4	1.5	3	4	Condominium	275.00000
4	1.5	3	4	House	171.94118
4	1.5	3	4	Houseboat	275.00000
4	1.5	3	4	Loft	225.00000
4	1.5	3	4	Townhouse	169.33333
4	1.5	3	5	Apartment	290.00000
4	1.5	3	5	Condominium	179.00000
4	1.5	3	5	House	179.33333
4	1.5	3	5	Townhouse	125.00000
4	1.5	3	6	Apartment	800.00000
4	1.5	3	6	Boat	301.00000
4	1.5	3	6	Loft	329.00000
4	1.5	3	6	Townhouse	251.66667
4	1.5	4	4	Apartment	199.90000
4	1.5	4	4	House	173.83333
4	1.5	4	4	Townhouse	254.14286
4	1.5	4	5	Apartment	700.00000
4	1.5	4	5	House	173.33333
4	1.5	4	6	House	187.50000
4	1.5	4	6	Loft	230.00000
4	1.5	4	6	Townhouse	470.00000
4	1.5	4	8	House	150.00000
4	2.0	0	6	House	185.00000
4	2.0	1	4	Apartment	187.50000
4	2.0	1	4	Hostel	90.00000
4	2.0	1	4	House	120.00000
4	2.0	1	4	Other	269.00000
4	2.0	2	4	Apartment	249.51429
4	2.0	2	4	Bed and breakfast	195.00000
4	2.0	2	4	Condominium	175.00000
4	2.0	2	4	House	208.75000
4	2.0	2	4	Houseboat	275.00000
4	2.0	2	4	Loft	400.00000
4	2.0	2	4	Townhouse	226.25000
4	2.0	2	5	Apartment	187.50000
4	2.0	2	6	Boat	300.00000
4	2.0	2	6	Loft	175.00000
4	2.0	3	4	Apartment	295.24000
4	2.0	3	4	Condominium	250.00000
4	2.0	3	4	House	230.03333
4	2.0	3	4	Houseboat	375.00000
4	2.0	3	4	Loft	150.00000
4	2.0	3	4	Townhouse	281.81818
4	2.0	3	5	Apartment	245.00000
4	2.0	3	5	House	177.50000
4	2.0	3	5	Loft	158.00000
4	2.0	3	5	Townhouse	150.00000
4	2.0	3	6	Apartment	329.00000
4	2.0	3	6	House	140.00000
4	2.0	3	6	Houseboat	595.00000
4	2.0	3	6	Townhouse	225.00000
4	2.0	3	7	House	249.00000
4	2.0	3	8	House	200.00000
4	2.0	4	4	Apartment	273.00000
4	2.0	4	4	Boat	375.00000
4	2.0	4	4	House	236.40000
4	2.0	4	4	Houseboat	420.00000
4	2.0	4	4	Loft	935.00000
4	2.0	4	4	Townhouse	272.50000
4	2.0	4	4	Villa	200.00000
4	2.0	4	6	Apartment	344.25000
4	2.0	4	6	House	306.25000
4	2.0	4	6	Houseboat	275.00000
4	2.0	4	6	Loft	249.00000
4	2.0	4	6	Townhouse	438.33333
4	2.0	4	6	Villa	280.00000
4	2.0	4	7	Townhouse	310.00000
4	2.0	4	8	Apartment	431.16667
4	2.0	4	8	House	304.50000
4	2.0	4	8	Loft	300.00000
4	2.0	5	8	Apartment	650.00000
4	2.5	2	4	Apartment	286.42857
4	2.5	2	4	Boat	170.00000
4	2.5	2	4	Serviced apartment	250.00000
4	2.5	3	4	Apartment	294.81818
4	2.5	3	4	House	244.66667
4	2.5	3	4	Loft	250.00000
4	2.5	3	4	Townhouse	235.00000
4	2.5	3	6	House	300.00000
4	2.5	4	4	Apartment	249.33333
4	2.5	4	4	House	190.00000
4	2.5	4	4	Houseboat	460.00000
4	2.5	4	4	Loft	200.00000
4	2.5	4	4	Townhouse	350.00000
4	2.5	4	5	House	237.50000
4	2.5	4	5	Townhouse	180.00000
4	2.5	4	6	Boat	250.00000
4	2.5	4	6	House	121.00000
4	2.5	4	6	Townhouse	185.00000
4	2.5	4	6	Villa	325.00000
4	2.5	8	4	House	300.00000
4	3.0	2	4	Apartment	220.00000
4	3.0	3	4	Apartment	430.00000
4	3.0	3	4	Houseboat	363.00000
4	3.0	3	8	Apartment	240.00000
4	3.0	3	8	House	285.00000
4	3.0	4	4	Apartment	295.00000
4	3.0	4	4	House	216.66667
4	3.0	4	4	Townhouse	290.75000
4	3.0	4	6	Apartment	343.00000
4	3.0	4	7	House	390.00000
4	3.0	4	8	Apartment	850.00000
4	3.5	3	4	Apartment	285.00000
4	3.5	3	4	Townhouse	412.50000
4	3.5	3	6	Townhouse	350.00000
4	3.5	4	4	Apartment	460.00000
4	3.5	4	4	House	150.00000
4	3.5	4	4	Townhouse	200.00000
4	4.0	3	8	House	2200.00000
4	4.0	4	8	Townhouse	500.00000
4	5.0	4	4	House	180.00000
4	10.0	1	4	Townhouse	120.00000
4	100.5	1	4	Apartment	65.00000
4	NA	1	4	Apartment	100.00000
5	0.0	1	3	Boat	30.00000
5	1.0	0	5	Boat	175.00000
5	1.0	1	3	Apartment	110.00000
5	1.0	1	4	Apartment	295.00000
5	1.0	1	5	Loft	120.00000
5	1.0	1	6	Apartment	274.50000
5	1.0	2	4	Apartment	237.00000
5	1.0	2	4	Bed and breakfast	90.00000
5	1.0	2	4	Condominium	150.00000
5	1.0	2	4	Houseboat	175.00000
5	1.0	2	5	Apartment	188.77778
5	1.0	2	5	Guesthouse	199.00000
5	1.0	2	6	Apartment	155.00000
5	1.0	2	7	House	186.00000
5	1.0	2	8	Apartment	240.00000
5	1.0	3	2	Townhouse	99.00000
5	1.0	3	4	Apartment	200.00000
5	1.0	3	4	House	171.16667
5	1.0	3	4	Houseboat	275.00000
5	1.0	3	4	Loft	250.00000
5	1.0	3	4	Townhouse	183.00000
5	1.0	3	4	Villa	240.00000
5	1.0	3	5	Apartment	186.78571
5	1.0	3	5	Boat	185.00000
5	1.0	3	5	Condominium	240.00000
5	1.0	3	5	House	183.18182
5	1.0	3	5	Townhouse	303.00000
5	1.0	3	6	Apartment	198.00000
5	1.0	3	6	House	240.00000
5	1.0	3	6	Houseboat	150.00000
5	1.0	3	6	Loft	290.00000
5	1.0	3	7	Apartment	165.00000
5	1.0	3	7	Bed and breakfast	70.00000
5	1.0	3	8	Apartment	119.00000
5	1.0	4	4	Apartment	186.25000
5	1.0	4	4	House	175.37500
5	1.0	4	4	Loft	475.00000
5	1.0	4	4	Townhouse	206.00000
5	1.0	4	4	Villa	217.50000
5	1.0	4	5	Apartment	245.00000
5	1.0	4	5	House	224.83333
5	1.0	4	5	Houseboat	195.00000
5	1.0	4	5	Townhouse	267.66667
5	1.0	4	6	Townhouse	395.00000
5	1.0	4	8	Apartment	100.00000
5	1.0	5	4	Apartment	179.00000
5	1.0	5	4	House	200.00000
5	1.0	5	5	Villa	200.00000
5	1.5	2	4	Apartment	194.33333
5	1.5	2	4	Bed and breakfast	120.00000
5	1.5	2	4	House	175.00000
5	1.5	2	4	Loft	213.00000
5	1.5	2	4	Serviced apartment	450.00000
5	1.5	2	5	Apartment	544.50000
5	1.5	2	5	Boat	150.00000
5	1.5	3	4	Apartment	213.75000
5	1.5	3	4	House	200.37500
5	1.5	3	4	Townhouse	192.50000
5	1.5	3	4	Villa	250.00000
5	1.5	3	5	Apartment	249.00000
5	1.5	3	5	House	211.00000
5	1.5	3	5	Townhouse	224.00000
5	1.5	3	6	Apartment	249.00000
5	1.5	3	6	Villa	275.00000
5	1.5	3	7	Apartment	250.00000
5	1.5	4	4	Apartment	255.00000
5	1.5	4	4	House	232.00000
5	1.5	4	4	Loft	350.00000
5	1.5	4	4	Townhouse	150.00000
5	1.5	4	5	House	200.00000
5	1.5	4	6	House	245.00000
5	1.5	4	8	Boat	499.00000
5	1.5	5	6	Houseboat	300.00000
5	1.5	5	10	Apartment	1000.00000
5	2.0	2	4	Apartment	300.00000
5	2.0	2	4	House	185.00000
5	2.0	2	9	Apartment	100.00000
5	2.0	2	10	Townhouse	500.00000
5	2.0	3	4	Apartment	242.50000
5	2.0	3	4	House	212.00000
5	2.0	3	5	Apartment	250.00000
5	2.0	3	5	House	192.71429
5	2.0	3	5	Loft	545.00000
5	2.0	3	5	Townhouse	250.00000
5	2.0	3	7	Apartment	277.00000
5	2.0	3	8	House	328.00000
5	2.0	4	4	Condominium	175.00000
5	2.0	4	4	House	264.00000
5	2.0	4	4	Other	299.00000
5	2.0	4	4	Townhouse	231.00000
5	2.0	4	4	Villa	525.00000
5	2.0	4	5	Apartment	400.00000
5	2.0	4	5	House	190.75000
5	2.0	4	5	Houseboat	150.00000
5	2.0	4	5	Townhouse	314.66667
5	2.0	4	6	Apartment	280.00000
5	2.0	4	6	House	207.50000
5	2.0	4	6	Townhouse	280.00000
5	2.0	4	8	Townhouse	333.33333
5	2.0	4	9	House	170.00000
5	2.0	5	4	Apartment	295.00000
5	2.0	5	4	House	300.00000
5	2.0	5	5	Townhouse	250.00000
5	2.0	5	6	Boat	340.00000
5	2.0	5	8	Villa	350.00000
5	2.0	5	9	House	385.00000
5	2.0	5	10	Apartment	300.00000
5	2.0	5	10	House	825.00000
5	2.5	3	3	Apartment	275.00000
5	2.5	3	4	Apartment	406.66667
5	2.5	3	4	House	232.33333
5	2.5	3	4	Townhouse	290.00000
5	2.5	3	5	House	150.00000
5	2.5	3	5	Townhouse	320.00000
5	2.5	3	6	Houseboat	350.00000
5	2.5	3	6	Loft	325.00000
5	2.5	4	4	Apartment	750.00000
5	2.5	4	5	Townhouse	275.00000
5	2.5	4	6	House	450.00000
5	2.5	4	7	House	210.00000
5	2.5	4	8	House	300.00000
5	2.5	5	4	House	185.00000
5	2.5	5	4	Townhouse	217.50000
5	2.5	5	8	House	249.00000
5	2.5	5	10	Townhouse	999.00000
5	3.0	3	5	House	195.00000
5	3.0	3	5	Townhouse	500.00000
5	3.0	3	7	Apartment	530.00000
5	3.0	4	4	Townhouse	250.00000
5	3.0	4	8	Apartment	1250.00000
5	3.5	5	10	House	475.00000
5	5.5	5	8	Townhouse	2000.00000
6	0.5	2	6	Boat	200.00000
6	0.5	3	6	Boat	179.00000
6	1.0	1	5	Apartment	130.00000
6	1.0	1	6	Boat	348.00000
6	1.0	1	6	Townhouse	95.00000
6	1.0	1	16	Bed and breakfast	3142.00000
6	1.0	2	4	Apartment	240.00000
6	1.0	2	4	Boat	220.00000
6	1.0	2	4	Houseboat	385.00000
6	1.0	2	4	Townhouse	87.00000
6	1.0	2	5	Apartment	299.00000
6	1.0	2	5	Condominium	220.00000
6	1.0	2	6	Apartment	204.00000
6	1.0	2	6	Bed and breakfast	99.00000
6	1.0	2	6	Boat	394.00000
6	1.0	2	6	Condominium	159.00000
6	1.0	2	6	House	87.00000
6	1.0	2	8	Apartment	165.00000
6	1.0	3	3	House	150.00000
6	1.0	3	4	Apartment	282.75000
6	1.0	3	4	Condominium	325.00000
6	1.0	3	4	House	202.57143
6	1.0	3	4	Townhouse	219.50000
6	1.0	3	5	Apartment	115.00000
6	1.0	3	5	Condominium	345.00000
6	1.0	3	5	House	160.00000
6	1.0	3	5	Loft	300.00000
6	1.0	3	5	Townhouse	200.00000
6	1.0	3	6	Apartment	277.84615
6	1.0	3	6	House	261.00000
6	1.0	4	4	Apartment	256.66667
6	1.0	4	4	Boat	280.00000
6	1.0	4	4	House	398.33333
6	1.0	4	4	Townhouse	209.60000
6	1.0	4	5	House	178.33333
6	1.0	4	5	Townhouse	317.50000
6	1.0	4	6	House	219.50000
6	1.0	4	7	Apartment	650.00000
6	1.0	4	8	House	150.00000
6	1.0	5	4	House	130.00000
6	1.0	5	5	House	220.00000
6	1.0	5	6	House	225.00000
6	1.0	6	6	House	209.00000
6	1.5	2	4	Apartment	135.00000
6	1.5	2	4	House	350.00000
6	1.5	2	6	Apartment	110.00000
6	1.5	2	6	Loft	110.00000
6	1.5	3	4	Apartment	311.80000
6	1.5	3	4	House	166.66667
6	1.5	3	4	Houseboat	300.00000
6	1.5	3	5	Apartment	386.50000
6	1.5	3	6	Apartment	311.12500
6	1.5	3	6	Condominium	200.00000
6	1.5	3	6	House	225.00000
6	1.5	3	6	Serviced apartment	750.00000
6	1.5	3	6	Villa	375.00000
6	1.5	3	7	Boat	198.00000
6	1.5	4	4	Apartment	250.00000
6	1.5	4	4	House	110.00000
6	1.5	4	4	Townhouse	227.50000
6	1.5	4	4	Villa	195.00000
6	1.5	4	5	Apartment	140.00000
6	1.5	4	6	Apartment	300.00000
6	1.5	4	6	House	199.00000
6	1.5	4	6	Serviced apartment	650.00000
6	1.5	4	8	House	175.00000
6	2.0	0	6	Townhouse	165.00000
6	2.0	2	6	Apartment	268.50000
6	2.0	2	6	Boat	337.00000
6	2.0	2	6	Condominium	400.00000
6	2.0	3	4	Apartment	237.00000
6	2.0	3	4	House	180.00000
6	2.0	3	4	Loft	524.50000
6	2.0	3	4	Townhouse	353.00000
6	2.0	3	5	House	230.00000
6	2.0	3	6	Apartment	514.25000
6	2.0	3	6	Boat	169.50000
6	2.0	3	6	House	324.50000
6	2.0	3	6	Houseboat	400.00000
6	2.0	3	6	Townhouse	233.00000
6	2.0	3	8	House	200.00000
6	2.0	4	4	Apartment	292.16667
6	2.0	4	4	Barn	400.00000
6	2.0	4	4	House	248.25000
6	2.0	4	4	Townhouse	199.80000
6	2.0	4	5	Apartment	181.66667
6	2.0	4	5	House	191.66667
6	2.0	4	5	Loft	350.00000
6	2.0	4	6	Apartment	300.00000
6	2.0	4	6	Boat	275.00000
6	2.0	4	6	House	321.80000
6	2.0	4	6	Houseboat	445.00000
6	2.0	4	6	Loft	250.00000
6	2.0	4	6	Townhouse	267.00000
6	2.0	4	8	Apartment	240.00000
6	2.0	4	8	House	300.00000
6	2.0	5	4	Cottage	170.00000
6	2.0	5	4	House	275.00000
6	2.0	5	4	Townhouse	350.00000
6	2.0	5	6	Apartment	225.00000
6	2.0	5	6	House	300.00000
6	2.0	5	6	Townhouse	425.00000
6	2.0	5	7	House	400.00000
6	2.0	5	8	Apartment	250.00000
6	2.0	5	8	House	731.33333
6	2.0	5	10	Apartment	900.00000
6	2.5	3	4	Apartment	512.50000
6	2.5	3	4	Townhouse	407.50000
6	2.5	3	6	House	258.00000
6	2.5	3	6	Loft	425.00000
6	2.5	3	6	Townhouse	185.00000
6	2.5	4	4	Townhouse	228.25000
6	2.5	4	6	Apartment	270.00000
6	2.5	4	6	House	325.00000
6	2.5	4	8	House	177.00000
6	2.5	5	4	Condominium	400.00000
6	2.5	5	4	House	500.00000
6	2.5	5	6	House	282.00000
6	3.0	3	6	House	500.00000
6	3.0	4	4	Apartment	550.00000
6	3.0	4	4	Townhouse	350.00000
6	3.0	4	5	Townhouse	425.00000
6	3.0	4	8	Apartment	550.00000
6	3.0	4	8	Loft	745.00000
6	3.0	5	6	House	795.00000
6	3.0	5	8	Apartment	550.00000
6	3.0	5	11	House	495.00000
6	3.5	3	4	Apartment	200.00000
6	3.5	5	8	Townhouse	680.00000
6	4.0	6	8	Townhouse	650.00000
7	1.0	0	6	Loft	295.00000
7	1.0	2	10	Apartment	196.00000
7	1.0	3	4	Apartment	250.00000
7	1.0	3	4	House	130.00000
7	1.0	3	7	Apartment	258.33333
7	1.0	3	12	Boat	420.00000
7	1.0	4	4	Apartment	130.00000
7	1.0	4	6	Apartment	287.50000
7	1.0	4	6	Boat	230.00000
7	1.0	4	6	House	345.00000
7	1.0	4	7	Apartment	395.00000
7	1.0	4	7	House	299.00000
7	1.0	5	11	Townhouse	300.00000
7	1.5	2	8	Apartment	190.00000
7	1.5	3	4	Boat	299.00000
7	1.5	3	6	House	175.00000
7	1.5	4	7	Apartment	649.50000
7	2.0	3	7	House	215.00000
7	2.0	4	4	Apartment	1032.50000
7	2.0	4	4	House	210.00000
7	2.0	4	6	House	250.00000
7	2.0	4	6	Villa	285.00000
7	2.0	4	7	Apartment	350.00000
7	2.0	4	7	House	270.00000
7	2.0	4	7	Houseboat	386.00000
7	2.0	4	7	Villa	175.00000
7	2.0	5	4	House	300.00000
7	2.0	5	4	Townhouse	200.00000
7	2.0	5	6	Houseboat	190.00000
7	2.0	5	7	Apartment	425.00000
7	2.0	5	7	Townhouse	500.00000
7	2.0	5	7	Villa	200.00000
7	2.5	3	8	Apartment	154.00000
7	2.5	4	7	Apartment	450.00000
7	2.5	7	4	House	124.00000
7	3.0	4	4	Loft	800.00000
7	3.0	4	7	House	225.00000
7	4.0	6	8	Apartment	174.00000
7	4.5	5	7	House	350.00000
8	1.0	3	6	Townhouse	200.00000
8	1.0	3	8	Apartment	201.00000
8	1.0	3	8	House	185.00000
8	1.0	4	4	House	500.00000
8	1.0	4	6	Apartment	173.00000
8	1.0	4	8	Apartment	200.00000
8	1.0	4	8	Boat	150.00000
8	1.0	4	8	House	190.00000
8	1.5	2	8	House	175.00000
8	1.5	3	8	Apartment	79.00000
8	1.5	4	4	Apartment	950.00000
8	1.5	4	8	House	195.00000
8	1.5	4	8	Townhouse	229.00000
8	1.5	5	4	House	275.00000
8	1.5	5	8	Boat	250.00000
8	2.0	3	4	Apartment	250.00000
8	2.0	4	4	House	599.00000
8	2.0	4	4	Loft	245.00000
8	2.0	4	6	House	445.00000
8	2.0	4	8	Apartment	260.00000
8	2.0	4	8	Villa	400.00000
8	2.0	5	6	Apartment	315.00000
8	2.0	5	6	Houseboat	303.00000
8	2.0	5	7	Townhouse	185.00000
8	2.5	3	7	Apartment	139.00000
8	2.5	4	8	Apartment	408.50000
8	3.0	1	8	Bed and breakfast	175.00000
8	3.0	1	8	Hostel	175.00000
8	3.0	3	4	House	295.00000
8	3.0	4	6	Townhouse	205.00000
8	3.0	4	7	Townhouse	499.00000
8	3.0	4	8	House	240.00000
8	3.0	4	8	Serviced apartment	623.50000
8	3.0	4	12	Apartment	400.00000
8	3.0	6	4	House	200.00000
8	3.5	5	7	Serviced apartment	700.00000
8	3.5	5	12	Loft	600.00000
8	4.0	4	8	House	400.00000
9	2.0	6	9	House	185.00000
9	3.0	6	4	House	385.00000
10	0.5	1	10	Houseboat	250.00000
10	1.0	2	12	House	225.00000
10	1.5	2	8	Apartment	350.00000
10	1.5	5	10	House	200.00000
10	2.0	5	8	Townhouse	275.00000
10	2.0	5	10	House	450.00000
10	2.5	5	10	Apartment	999.00000
10	3.0	3	9	Apartment	795.00000
10	3.0	5	8	House	376.00000
10	3.5	5	10	House	299.00000
10	3.5	6	4	House	511.00000
10	4.0	6	6	Apartment	314.00000
11	2.0	4	16	Apartment	480.00000
11	2.0	5	10	House	700.00000
11	2.0	5	11	Serviced apartment	900.00000
12	1.0	1	12	Boat	550.00000
12	1.0	1	14	Loft	800.00000
12	1.0	2	12	House	200.00000
12	1.0	6	12	Boat	383.00000
12	1.5	3	12	Apartment	650.00000
12	1.5	5	12	Apartment	650.00000
12	2.0	4	12	Townhouse	345.00000
12	2.0	4	16	Apartment	480.00000
12	2.0	6	12	Boat	450.00000
12	2.5	6	4	Townhouse	450.00000
12	3.0	3	12	Bungalow	280.00000
12	3.0	6	12	Townhouse	500.00000
14	2.0	3	14	Boat	425.00000
14	2.0	7	14	Boat	575.00000
15	2.0	3	16	House	425.00000
16	1.0	5	16	Boat	490.00000
16	1.0	6	16	Boat	398.00000
16	2.0	5	16	Boat	600.00000
16	2.0	5	16	Houseboat	700.00000
16	2.0	6	16	Boat	474.00000
16	2.0	9	16	Boat	600.00000
16	2.0	10	16	Boat	382.00000
16	3.5	7	16	Apartment	1600.00000
16	4.0	10	16	Boat	597.50000
16	8.0	1	16	Bed and breakfast	106.00000
16	8.0	10	16	Boat	33.00000
17	5.0	5	17	Hotel	210.00000
18	1.0	6	16	Boat	495.00000
18	1.5	8	16	Boat	465.00000
18	2.0	8	16	Boat	440.00000
20	2.0	8	16	Boat	495.00000
22	2.0	11	16	Boat	447.00000
26	3.0	10	16	Boat	800.00000
32	4.0	12	16	Boat	286.66667

5) Amenities

Rooms amenities are also important. Whether your unit comes with a pool, parking space, fireplace, kitchen, or close to waterfront could largely change tourists’ staying experience. Hence, we look at all the individual amenities and their effect on price.

#pool
details.sf <- details.sf %>%
  mutate(pool = ifelse(str_detect(amenities, "Pool"), "Yes", "No"))

#Paid parking off premises
details.sf <- details.sf %>%
  mutate(parking = ifelse(str_detect(amenities, "Paid parking off premises"), 
                          "Yes", "No"))

#Indoor fireplace
details.sf <- details.sf %>%
  mutate(fireplace = ifelse(str_detect(amenities, "Indoor fireplace"), 
                          "Yes", "No"))

#Waterfront
details.sf <- details.sf %>%
  mutate(waterfront = ifelse(str_detect(amenities, "Waterfront"), 
                          "Yes", "No"))

#Kitchen
details.sf <- details.sf %>%
  mutate(kitchen = ifelse(str_detect(amenities, "Kitchen"), 
                          "Yes", "No"))

#Air conditioning
details.sf <- details.sf %>%
  mutate(AC = ifelse(str_detect(amenities, "Air conditioning"), 
                          "Yes", "No"))

amenitie_vars <- c('fireplace','waterfront','pool','parking','kitchen','AC')
plotList <- list()

for (i in amenitie_vars){
plotList[[i]] <- 
  details.sf %>%st_drop_geometry() %>% 
  dplyr::select(price,i) %>%
  filter(price<500) %>% 
  gather(Variable, value, -price) %>%
    ggplot(aes(value, price, fill=value)) + 
      geom_boxplot() +
      scale_fill_manual(values = c("#FFDD94", "#CCABDB")) +
      labs(x="Amenity Status", y="Price",
           title = i) +
      theme(legend.position = "none") +
  plotTheme()
}

do.call(grid.arrange,c(plotList, ncol = 3, top = "Price w/ vs. w/o Amenities"))

From the box plot above, we can tell that fireplace, waterfront, pool, and kitchen seem to have a decent influence on price, while parking and AC have subtle effect. Since Amsterdam is usually cool in the summer and chilly in winter, it makes sense why AC maybe less appealing than fireplace. We

Aggregated Amenities:

In addition to each amenities listed above, we also included a feature named ‘amenities.number’ to explore the aggregated effect of amenities. Sometimes, regardless of specific types of amenities, simply the fact of having multiple amenities may increase the price visitors are willing to pay for an Airbnb.

New Features & Engineering

In addition to internal features that we examined earlier, we also want to take some external factors that are critical for tourists into considerations – the neighborhood (area), distance to public transit stops, and distance to some tourists attractions (e.g. museums) and external amenities (supermarkets and convenience shops). As tourists can benefit a lot from close proximity to sites they want to visit or amenities, the price may go up for those units that are closer to those places.

New Feature 1: Spatial Lag & Neighborhood Effect

There is a subtle difference between neighborhood effect – spatial lag takes the price / revenue of surrounding areas and use the spatial lag revenue as a factor for pricing; neighborhood effect considers the spatial aspects and use neighborhood as a factor, since the location of a unit may silently speak for many other factors that may impact the price. Here we are considering both.

Neighborhood Effect

To account for neighborhood effect, we used the smaller scale 481 neighborhood system instead of the larger districts to increase precision.

details.sf.neighbor <- st_intersection(details.sf,zipcode4) %>% 
  dplyr::select(id, Postcode4) %>% 
  st_drop_geometry()

details.sf <- left_join(details.sf,details.sf.neighbor,by='id')

neighbor2 <- neighbor2 %>% 
  st_transform(st_crs(listings.sf))

listing.sf.neighbor2 <- st_intersection(listings.sf,neighbor2) %>% 
  dplyr::select(id, Buurt,Buurt_code) %>% 
  mutate(count=1) %>% st_drop_geometry()

listing.sf.neighbor2$id <- as.character(listing.sf.neighbor2$id)

detail.sf <- left_join(details.sf,listing.sf.neighbor2, by = "id")

Spatial Lag

To calculate spatial lag, we used revenue instead of price because while price may be a subjective decision by the host, revenue speaks more about the market and how this area is doing in terms of profitability. Whether or not a host is likely to profit may impact their business decisions.

occupancy3 <- occupancy %>%
  dplyr::rename(id = listing_id) 
  
occupancy3 <-  
  left_join(occupancy3, st_drop_geometry(details.sf), by = "id")

listing_panel <- listing_panel%>%
   dplyr::rename(id = listing_id)

revenue_panel <- merge(occupancy3,listing_panel[c("id","month","each_month_price")],by=c("id","month")) %>%
  mutate(revenue = each_month_price*monthly_occupancy)

annualrevenue <- revenue_panel %>%
  group_by(id) %>%
  dplyr::summarise(annual_revenue = sum(revenue))


annualrevenue<- left_join(details.sf, annualrevenue,by="id")%>%
    filter(!id %in% no_price)%>%
    mutate(bathrooms = as.numeric(bathrooms),
         bedrooms = as.numeric(bedrooms))

annualrevenue<- annualrevenue %>%
  drop_na(annual_revenue) %>%
  filter(annual_revenue <= 500000)

coords <- st_coordinates(annualrevenue) 

neighborList <- knn2nb(knearneigh(coords, 5))

spatialWeights <- nb2listw(neighborList, style="W")


annualrevenue$lagRevenue <- lag.listw(spatialWeights, annualrevenue$annual_revenue)

ggplot(annualrevenue)+
  geom_point(aes(x = price, y = annual_revenue), alpha = 0.26)+
  geom_smooth(aes(x = price, y =annual_revenue), method = "lm", se= FALSE, color = "orange")+
  labs(title="Price as a function of lagRevenue", y='Price', x = 'Lag Annual Revenue') +
  plotTheme()

The graph above shows price as function of lag annual revenue. We can recognize a clear positive correlation from this graph – the more money neighboring Airbnb makes, the higher an Airbnb is tend to price since they may have the confidence of having sufficient demand. We therefore also included lagRevenue into our model.

New Feature 2: Access to Public Transit

The second set of features is the proximity to public transit. We first plotted the existing metro and tram stations in the city. We can see a pretty clear pattern of higher density of Aibnb near stations. But, does it matter when it comes to pricing?

This also hints that if the government are to generate new policies for encouraging development in other areas, including directing more Airbnb hosts to rent out units in peripheral areas, they may need to solve the public transit accessibility issue first.

listings$longitude = as.numeric(listings$longitude)
listings$latitude = as.numeric(listings$latitude)
listings.sf<- listings %>% 
  st_as_sf(coords = c( "longitude","latitude"), crs = 4326, agr = "constant")

st_crs(metro) <- st_crs(listings.sf)

st_c <- st_coordinates

details.sf.c <- details.sf %>%
  st_transform('ESRI:102013') 

metro.c <- metro%>%
  st_transform('ESRI:102013') 

ggplot()+
  geom_sf(data = neighbor2, aes(fill=q5(airbnb.number)),color='transparent') +
  scale_fill_manual(values = Pink,
                     labels = qBr(neighbor2,'airbnb.number'), name = 'Count of Airbnb') +
  geom_sf(data=metro.c$geometry, aes(color=metro.c$Modaliteit)) +
  scale_color_manual(values=c(palette2[1], '#7171C6'), name = 'Type of Transit') +
  labs(title = "Public Transit & Number of Airbnb", subtitle = "Amsterdam, NL; 2018") +
  mapTheme()

#multi-ring buffer 
Metro_MRB <- multipleRingBuffer(st_union(metro.c), 6500, 500)

#USE FOR REG: MRB Spatial Join 
details.sf.c.MetroRings <-
  st_join(dplyr::select(details.sf.c, id),
          Metro_MRB) %>%
  st_drop_geometry() %>% 
  left_join(details.sf.c, annualrevenue,
            by="id") %>%
  st_sf()%>%
  mutate(metro.dist = distance/1000)

#Adding the variable to the data set
details.sf <- 
  left_join(details.sf, details.sf.c.MetroRings %>% st_drop_geometry() %>% dplyr::select(metro.dist, id), by="id")

#FOR VIS ONLY: Creating a summary table of the ring data 
details.Rings.Summary <- details.sf.c.MetroRings %>%
  dplyr::select(metro.dist, price)%>%
  st_drop_geometry()%>%
  group_by(metro.dist)%>%
  dplyr::summarize(Price=median(price, na.rm=T))

#FOR VIS ONLY: Line Plot of the summary table 
ggplot(data=details.Rings.Summary, aes(x=metro.dist, y=Price),)+
  geom_line(size=1.5)+
  geom_point(size = 3)+  
  labs(title = "Price as a Function of Distance from a Metro Station", 
       subtitle = "Amsterdam Airbnb Data, 2018", 
       x="Distance from a Metro Station (KM)",
       y="Price (2018 Euro)")+
  theme(plot.title = element_text(size=18))

The plot above shows an interesting pattern of the relationship between price and distance from metro station. we can see a declining trend from the closest to station to around 4.5 km, then a striking increase in price. This could be due to the fact that after 4.5 it falls into the close proximity to another station, hence the price is risen again. The dramatic ups and downs also signals that it might be a valid factor for pricing.

annualrevenue <-
  annualrevenue %>%
  left_join(details.sf.c.MetroRings %>% st_drop_geometry %>% dplyr::select(metro.dist, id), by = 'id')

New Feature 3: Proximity to Tourist Amenities

The third set of features we look at are tourist amenities, including UNESCO sites, parks, tourist attractions, museums, malls, convenience shops, plazas, and supermarkets. They are either possible incentives for tourists or just amenities for convenience / daily shopping. Given the fact that Airbnb is unlike hotel where all kinds of services are provided, tourists staying at Airbnb may be visiting shopping facilities more frequently to do last-minute shopping.

unesco <-
  #st_read('UnescoWerelderfgoed_region.shp') %>%
  st_read('geojson_lnglat.json') %>%
  st_transform('ESRI:102013')

parks <- st_read('parks.json') %>% 
  st_transform('ESRI:102013')

attraction <- st_read('amsterdam-attraction.csv') %>% 
  filter(lat != 'attraction')
attraction <- st_as_sf(attraction,coords = c('lng','lat'),crs = 4326) %>% 
  st_transform('ESRI:102013')

parks <- attraction %>% 
  filter(subCategory == 'Park')

museum <- attraction %>% 
  filter(subCategory == 'Museum')

#supermarkets
supermarkets <- getbb('Amsterdam') %>% 
  opq() %>% 
  add_osm_feature('shop','supermarket') %>% 
  osmdata_sf() 

supermarkets <- supermarkets$osm_points %>% 
  dplyr::select(geometry)%>% 
  st_transform('ESRI:102013')%>% 
  dplyr::select(geometry) %>% 
  mutate(Legend = "supermarkets")

convenienceshop <- getbb('Amsterdam') %>% 
  opq() %>% 
  add_osm_feature('shop','convenience') %>% 
  osmdata_sf() 

convenienceshop <- convenienceshop$osm_points %>% 
  dplyr::select(geometry)%>% 
  st_transform('ESRI:102013')%>% 
  dplyr::select(geometry) %>% 
  mutate(Legend = "convenienceshop")

mall <- getbb('Amsterdam') %>% 
  opq() %>% 
  add_osm_feature('shop','mall') %>% 
  osmdata_sf() 

mall<- mall$osm_points %>% 
  dplyr::select(geometry)%>% 
  st_transform('ESRI:102013') %>% 
  dplyr::select(geometry) %>% 
  mutate(Legend = "mall")

attraction <- st_as_sf(attraction,coords = c('lng','lat'),crs = 4326) %>% 
  st_transform('ESRI:102013')


plaza <- attraction %>% 
  filter(subCategory == 'Plaza') 


beach <- attraction %>% 
  filter(subCategory == 'Beach') 


nightclub <- attraction %>% 
  filter(subCategory == 'Nightclub') 

dist <- 1000

details.sf.c.attractions <- details.sf %>%
  dplyr::select(id)

#malls
mallBuffer <- st_union(st_buffer(mall, dist))%>%
    st_sf()

details.sf.c.attractions <- 
  rbind(
    details.sf.c[mallBuffer,] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(mall.close = "y"),
    details.sf.c[mallBuffer, op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(mall.close = "n"))
  
#supermarkets
supermarketsBuffer <- st_union(st_buffer(supermarkets, dist))%>%
    st_sf()

details.sf.c.attractions <- 
  rbind(
    details.sf.c[supermarketsBuffer,] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(supermarkets.close = "y"),
    details.sf.c[supermarketsBuffer, op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(supermarkets.close = "n"))

#dist.convenienceshop

details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(convenienceshop, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(convenienceshop.close = "y"),
    details.sf.c[st_union(st_buffer(convenienceshop, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(convenienceshop.close = "n"))

#dist.museum
details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(museum, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(museum.close = "y"),
    details.sf.c[st_union(st_buffer(museum, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(museum.close = "n"))

#dist.plaza
details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(plaza, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(plaza.close = "y"),
    details.sf.c[st_union(st_buffer(plaza, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(plaza.close = "n"))

#dist.nightclub
details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(nightclub, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(nightclub.close = "y"),
    details.sf.c[st_union(st_buffer(nightclub, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(nightclub.close = "n"))

#dist.beach
details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(beach, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(beach.close = "y"),
    details.sf.c[st_union(st_buffer(beach, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(beach.close = "n"))

#dist.parks
details.sf.c.attractions <- 
  rbind(
    details.sf.c[st_union(st_buffer(parks, dist)),] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(parks.close = "y"),
    details.sf.c[st_union(st_buffer(parks, dist)), op = st_disjoint] %>%
      st_drop_geometry() %>%
      left_join(details.sf.c.attractions) %>%
      st_sf() %>%
      mutate(parks.close = "n"))

details.sf.attractions <-
  left_join(details.sf, st_drop_geometry(details.sf.c.attractions), by="id")

table.distance <- details.sf.attractions %>% 
  st_drop_geometry() %>% 
  dplyr::select(price.y, museum.close,plaza.close,nightclub.close,beach.close,
                  supermarkets.close, parks.close, mall.close, convenienceshop.close)

table.distance %>%
  gather(Variable, value, -price.y) %>%
    ggplot(aes(value, price.y, fill=value)) + 
      geom_bar(position = "dodge", stat = "summary", fun = "mean") + 
      facet_wrap(~Variable, scales = "free") +
      scale_fill_manual(values = palette2) +
      labs(x="Within 1KM?", y="Average Price", 
           title = "Feature Associations with Proximity to Attractions",
           subtitle = "Amsterdam Airbnbs, 2018") +
      theme(legend.position = "none")

Similar to internal amenities, we drew box plots for external amenities. We compared price of Airbnb that has particular amenities within the 1km walkable distance, versus without those amenities in close proximity. We can see that apart from beach, nightclubs, and parks, all five other factors seem to make a difference when it comes to determining Airbnb price.

annualrevenue <-
  annualrevenue %>%
  left_join(details.sf.c.attractions %>% st_drop_geometry %>% dplyr::select(id, convenienceshop.close, mall.close, museum.close, plaza.close, supermarkets.close), by = 'id')

Feature Review: Correlation Matrix

The correlation matrix below is a quick check-up on whether our factors are correlated to each other. If they are, we will choose one to include in model and the other(s) to disregard. Apart from beds and bedrooms, all other factors seem to be fine to be included in the model simultaneously.

numericVars <- 
  select_if(st_drop_geometry(annualrevenue), is.numeric) %>% na.omit() %>% 
  dplyr::select(annual_revenue,price,beds, bedrooms, bathrooms, 
                              minimum_nights,
                              amenities.number, metro.dist)

ggcorrplot(
  round(cor(numericVars), 1), 
  p.mat = cor_pmat(numericVars),
  colors = c("#4757a2", "white", "#E46B45"),
  type="lower",
  insig = "blank") +  
    labs(title = "Correlation across numeric variables") 

Regression Model

We are using linear regression OLS model for this machine learning task.

Data Split: Testing & Training

To train and test our model, we are splitting the existing data into 60% for training and 40% for testing. Since several neighborhoods only have 1 Airbnb in the area, we are keeping those data in the training set to avoid errors. However, we are aware that this results more neighbors having NAs.

annualrevenue <- merge(annualrevenue,listing.sf.neighbor2[c("id", "Buurt")], by = "id")

#Split training and test set
annualrevenue <- annualrevenue%>% st_drop_geometry()
set.seed(31497)

inTrain <- caret::createDataPartition(
  y =  paste(annualrevenue$Buurt),
  p = .6, list = FALSE)

annualrevenue.training <- st_drop_geometry(annualrevenue)[inTrain,]
annualrevenue.test     <- st_drop_geometry(annualrevenue)[-inTrain,]

Regression Results

The long table below shows the regression results from our model. The model has \(R^2\) of 0.426 and most of the features we included appear to be significant. Although 0.426 may not be a high model for price prediction, we want to acknowledge the fact that Airbnb price may be more subjective, fluctuating, and unstable than house prices due to many unforeseeable factors.

reg.annualrevenue <- lm(price ~ ., data = st_drop_geometry(annualrevenue) %>% 
             dplyr::select(price, beds, bedrooms, bathrooms, accommodates,
                              pool, parking, kitchen, AC, fireplace,
                              room_type,property_type,bed_type, Buurt,
                              minimum_nights, convenienceshop.close, mall.close, museum.close,
                              plaza.close, supermarkets.close,
                              amenities.number,lagRevenue, metro.dist)) 

stargazer(reg.annualrevenue,
          type = "text",
          title ="Regression Output",
          single.row = TRUE,
          out.header = TRUE)

## 
## Regression Output
## =======================================================================
##                                                 Dependent variable:    
##                                             ---------------------------
##                                                        price           
## -----------------------------------------------------------------------
## beds                                               0.553 (0.779)       
## bedrooms                                         21.595*** (1.066)     
## bathrooms                                        4.544*** (0.703)      
## accommodates                                     22.155*** (0.754)     
## poolYes                                           15.749* (8.801)      
## parkingYes                                       9.563*** (1.447)      
## kitchenYes                                        4.264* (2.301)       
## ACYes                                            22.615*** (2.752)     
## fireplaceYes                                     33.248*** (2.498)     
## room_typePrivate room                           -34.166*** (1.650)     
## room_typeShared room                            -42.477*** (9.784)     
## property_typeApartment                          -37.675*** (9.809)     
## property_typeBarn                                -29.573 (40.038)      
## property_typeBed and breakfast                  -28.035*** (10.546)    
## property_typeBoat                               -31.779*** (10.661)    
## property_typeBoutique hotel                      -12.533 (17.841)      
## property_typeBungalow                            -22.145 (29.906)      
## property_typeCabin                               -21.923 (23.194)      
## property_typeCamper/RV                           -90.810 (55.749)      
## property_typeCampsite                            -69.031 (57.643)      
## property_typeCasa particular (Cuba)               -4.534 (35.637)      
## property_typeCastle                              -37.605 (77.178)      
## property_typeChalet                              -34.975 (45.723)      
## property_typeCondominium                        -36.010*** (10.696)    
## property_typeCottage                             -28.512 (25.435)      
## property_typeEarth house                         -52.881 (76.736)      
## property_typeGuest suite                        -40.507*** (11.684)    
## property_typeGuesthouse                          -22.132 (15.862)      
## property_typeHostel                              -29.629 (41.344)      
## property_typeHotel                                -3.484 (30.731)      
## property_typeHouse                              -30.511*** (10.081)    
## property_typeHouseboat                            -6.386 (11.136)      
## property_typeLighthouse                         667.496*** (79.423)    
## property_typeLoft                                 -3.672 (10.514)      
## property_typeNature lodge                         26.590 (81.860)      
## property_typeOther                              -35.733** (14.762)     
## property_typeServiced apartment                  29.532** (13.812)     
## property_typeTent                                -43.168 (80.574)      
## property_typeTiny house                          -36.244 (35.847)      
## property_typeTownhouse                          -26.219** (10.295)     
## property_typeVilla                               -13.132 (17.276)      
## bed_typeCouch                                    -26.797 (43.084)      
## bed_typeFuton                                     -7.595 (22.902)      
## bed_typePull-out Sofa                             0.927 (20.872)       
## bed_typeReal Bed                                  10.019 (19.847)      
## BuurtAalsmeerwegbuurt West                      35.143*** (10.604)     
## BuurtAlexanderplein e.o.                         42.658* (24.094)      
## BuurtAmstel III deel A/B Noord                    -7.352 (76.806)      
## BuurtAmstelglorie                                 -0.891 (25.590)      
## BuurtAmstelkwartier Noord                        32.510** (13.989)     
## BuurtAmstelkwartier West                          -2.734 (35.511)      
## BuurtAmstelkwartier Zuid                          33.744 (76.823)      
## BuurtAmstelpark                                  -71.799 (77.047)      
## BuurtAmstelveldbuurt                            126.564*** (13.380)    
## BuurtAmsterdamse Bos                              19.128 (26.876)      
## BuurtAmsterdamse Poort                            -9.780 (24.864)      
## BuurtAndreasterrein                               13.350 (23.928)      
## BuurtAnjeliersbuurt Noord                       57.773*** (12.510)     
## BuurtAnjeliersbuurt Zuid                        54.526*** (11.949)     
## BuurtArchitectenbuurt                             3.238 (18.204)       
## BuurtBalboaplein e.o.                             15.673 (11.988)      
## BuurtBanne Noordoost                             -22.560 (23.110)      
## BuurtBanne Noordwest                              16.230 (25.795)      
## BuurtBanne Zuidoost                              -19.975 (18.654)      
## BuurtBanne Zuidwest                               2.862 (21.075)       
## BuurtBanpleinbuurt                               42.261** (21.480)     
## BuurtBedrijvencentrum Osdorp                      36.418 (76.491)      
## BuurtBedrijvencentrum Westerkwartier             -24.935 (32.409)      
## BuurtBedrijvengebied Cruquiusweg                  3.315 (39.166)       
## BuurtBedrijvengebied Veelaan                      16.839 (76.510)      
## BuurtBedrijvengebied Zeeburgerkade               84.986* (44.902)      
## BuurtBedrijvenpark Lutkemeer                     -103.888 (77.467)     
## BuurtBedrijventerrein Hamerstraat                 -4.208 (25.927)      
## BuurtBedrijventerrein Landlust                    15.188 (16.131)      
## BuurtBedrijventerrein Schinkel                    21.193 (16.474)      
## BuurtBeethovenbuurt                             55.193*** (21.216)     
## BuurtBegijnhofbuurt                             117.235*** (17.552)    
## BuurtBelgiëplein e.o.                             0.776 (35.267)       
## BuurtBellamybuurt Noord                         32.077*** (11.585)     
## BuurtBellamybuurt Zuid                           26.075** (10.877)     
## BuurtBertelmanpleinbuurt                          21.082 (20.081)      
## BuurtBetondorp                                   -23.166 (20.862)      
## BuurtBG-terrein e.o.                            59.245*** (15.166)     
## BuurtBijlmermuseum Noord                          37.116 (35.804)      
## BuurtBijlmermuseum Zuid                          -50.627 (35.775)      
## BuurtBijlmerpark Oost                             -5.176 (54.717)      
## BuurtBlauwe Zand                                 -17.971 (16.018)      
## BuurtBloemenbuurt Noord                           -2.490 (16.294)      
## BuurtBloemenbuurt Zuid                           -12.445 (16.044)      
## BuurtBloemgrachtbuurt                           64.940*** (11.472)     
## BuurtBorgerbuurt                                33.633*** (11.700)     
## BuurtBorneo                                       3.687 (14.120)       
## BuurtBosleeuw                                     -0.833 (11.513)      
## BuurtBretten Oost                                 63.320 (76.814)      
## BuurtBuiksloterbreek                             -10.179 (28.781)      
## BuurtBuiksloterdijk West                          -4.209 (35.731)      
## BuurtBuiksloterham                                22.447 (32.598)      
## BuurtBuikslotermeer Noord                        -22.314 (25.913)      
## BuurtBuikslotermeerplein                          2.998 (23.198)       
## BuurtBuitenveldert Midden Zuid                   33.960** (16.100)     
## BuurtBuitenveldert Oost Midden                    8.646 (18.357)       
## BuurtBuitenveldert West Midden                   81.099** (35.860)     
## BuurtBuitenveldert Zuidoost                       -0.105 (17.897)      
## BuurtBuitenveldert Zuidwest                       2.636 (15.970)       
## BuurtBurgemeester Tellegenbuurt Oost            42.951*** (13.872)     
## BuurtBurgemeester Tellegenbuurt West             30.002** (14.444)     
## BuurtBurgwallen Oost                            132.658*** (13.129)    
## BuurtBuurt 10                                    -33.986 (30.284)      
## BuurtBuurt 2                                      -3.997 (21.681)      
## BuurtBuurt 3                                     -20.596 (14.300)      
## BuurtBuurt 4 Oost                                 -0.237 (22.640)      
## BuurtBuurt 5 Noord                               -19.185 (32.533)      
## BuurtBuurt 5 Zuid                                -16.319 (24.444)      
## BuurtBuurt 6                                      11.239 (38.780)      
## BuurtBuurt 7                                      10.404 (34.916)      
## BuurtBuurt 8                                      0.677 (24.243)       
## BuurtBuurt 9                                     -54.151 (44.566)      
## BuurtBuyskade e.o.                                17.604 (12.147)      
## BuurtCalandlaan/Lelylaan                         -14.475 (26.851)      
## BuurtCentrumeiland                                23.713 (76.744)      
## BuurtCircus/Kermisbuurt                          -62.709 (47.663)      
## BuurtCoenhaven/Mercuriushaven                     38.978 (76.894)      
## BuurtColumbusplein e.o.                           17.873 (11.601)      
## BuurtConcertgebouwbuurt                         62.511*** (15.124)     
## BuurtCornelis Douwesterrein                       11.140 (55.841)      
## BuurtCornelis Schuytbuurt                       89.987*** (13.469)     
## BuurtCornelis Troostbuurt                       36.467*** (10.813)     
## BuurtCremerbuurt Oost                           41.714*** (11.358)     
## BuurtCremerbuurt West                           27.583*** (10.287)     
## BuurtCzaar Peterbuurt                             5.140 (13.933)       
## BuurtD-buurt                                      -2.483 (31.090)      
## BuurtDa Costabuurt Noord                        49.643*** (10.461)     
## BuurtDa Costabuurt Zuid                         55.354*** (11.598)     
## BuurtDapperbuurt Noord                           25.849** (11.398)     
## BuurtDapperbuurt Zuid                             12.369 (11.625)      
## BuurtDe Aker Oost                                 -5.281 (16.205)      
## BuurtDe Aker West                               -45.290** (22.895)     
## BuurtDe Bongerd                                  -18.050 (17.858)      
## BuurtDe Eenhoorn                                  13.326 (19.758)      
## BuurtDe Kleine Wereld                            -14.556 (28.514)      
## BuurtDe Klenckebuurt                              49.772 (39.176)      
## BuurtDe Omval                                    41.803* (25.042)      
## BuurtDe Punt                                      1.612 (20.281)       
## BuurtDe Wester Quartier                         41.068*** (13.754)     
## BuurtDe Wetbuurt                                  -5.922 (18.107)      
## BuurtDe Wittenbuurt Noord                       39.645*** (13.968)     
## BuurtDe Wittenbuurt Zuid                          22.435 (20.261)      
## BuurtDelflandpleinbuurt Oost                      16.932 (28.012)      
## BuurtDelflandpleinbuurt West                      9.075 (16.126)       
## BuurtDen Texbuurt                               64.358*** (14.564)     
## BuurtDiamantbuurt                                 15.154 (12.765)      
## BuurtDiepenbrockbuurt                             37.977 (35.253)      
## BuurtDon Bosco                                    9.422 (13.824)       
## BuurtDorp Driemond                                4.599 (38.102)       
## BuurtDorp Sloten                                  -4.319 (23.412)      
## BuurtDriehoekbuurt                              51.107*** (11.854)     
## BuurtDuivelseiland                              63.847*** (14.525)     
## BuurtDurgerdam                                   -18.683 (24.346)      
## BuurtE-buurt                                     -14.550 (26.618)      
## BuurtEcowijk                                      12.248 (18.174)      
## BuurtEendrachtspark                              -12.432 (76.556)      
## BuurtElandsgrachtbuurt                          76.607*** (11.300)     
## BuurtElzenhagen Noord                           -30.912** (15.413)     
## BuurtElzenhagen Zuid                              13.170 (76.587)      
## BuurtEmanuel van Meterenbuurt                    -12.440 (18.810)      
## BuurtEntrepot-Noordwest                           2.098 (18.430)       
## BuurtErasmusparkbuurt Oost                        -0.855 (13.980)      
## BuurtErasmusparkbuurt West                        4.838 (10.971)       
## BuurtF-buurt                                     -12.408 (23.723)      
## BuurtFannius Scholtenbuurt                      29.590*** (11.030)     
## BuurtFelix Meritisbuurt                         93.404*** (12.472)     
## BuurtFilips van Almondekwartier                   21.730 (13.752)      
## BuurtFlevopark                                    4.273 (39.199)       
## BuurtFrankendael                                  6.655 (27.035)       
## BuurtFrans Halsbuurt                            35.892*** (11.737)     
## BuurtFrederik Hendrikbuurt Noord                42.887*** (10.746)     
## BuurtFrederik Hendrikbuurt Zuidoost             43.797*** (11.042)     
## BuurtFrederik Hendrikbuurt Zuidwest              23.461* (14.169)      
## BuurtFrederikspleinbuurt                        51.355*** (14.135)     
## BuurtG-buurt Noord                                -4.115 (35.838)      
## BuurtG-buurt Oost                                 -0.201 (22.728)      
## BuurtG-buurt West                                -17.985 (21.250)      
## BuurtGaasperdam Noord                            -20.529 (26.076)      
## BuurtGaasperdam Zuid                            -72.633** (35.403)     
## BuurtGaasperpark                                  35.262 (46.883)      
## BuurtGaasperplas                                 -17.549 (40.470)      
## BuurtGein Noordoost                              -11.997 (21.083)      
## BuurtGein Noordwest                              -25.475 (30.374)      
## BuurtGein Zuidwest                                5.333 (54.608)       
## BuurtGein Zuioost                                -44.856 (28.689)      
## BuurtGelderlandpleinbuurt                         0.861 (14.579)       
## BuurtGerard Doubuurt                            56.063*** (11.816)     
## BuurtGeuzenhofbuurt                               18.096 (12.169)      
## BuurtGibraltarbuurt                               2.262 (11.913)       
## BuurtGouden Bocht                               82.485*** (20.082)     
## BuurtGroenmarktkadebuurt                        41.381*** (15.872)     
## BuurtGrunder/Koningshoef                          38.471 (26.588)      
## BuurtHaarlemmerbuurt Oost                       63.105*** (13.646)     
## BuurtHaarlemmerbuurt West                       46.023*** (12.272)     
## BuurtHakfort/Huigenbos                           -31.004 (33.124)      
## BuurtHarmoniehofbuurt                            53.500** (25.994)     
## BuurtHaveneiland Noord                            -4.536 (19.420)      
## BuurtHaveneiland Noordoost                        1.689 (14.176)       
## BuurtHaveneiland Noordwest                        -2.227 (17.525)      
## BuurtHaveneiland Oost                             -0.716 (13.827)      
## BuurtHaveneiland Zuidwest/Rieteiland West         -8.744 (14.549)      
## BuurtHelmersbuurt Oost                          44.404*** (11.693)     
## BuurtHemelrijk                                  50.912*** (14.572)     
## BuurtHemonybuurt                                49.384*** (11.273)     
## BuurtHercules Seghersbuurt                      53.134*** (12.052)     
## BuurtHet Funen                                    18.021 (17.656)      
## BuurtHiltonbuurt                                 82.887** (39.047)     
## BuurtHolendrecht Oost                            -44.064 (32.802)      
## BuurtHolendrecht West                           205.055*** (76.640)    
## BuurtHolysloot                                   -22.619 (38.805)      
## BuurtHondecoeterbuurt                           52.824*** (15.023)     
## BuurtHoptille                                    -15.954 (33.171)      
## BuurtHouthavens Oost                             38.354* (19.973)      
## BuurtHouthavens West                              22.414 (17.319)      
## BuurtIJplein e.o.                                 8.815 (15.117)       
## BuurtIJsbaanpad e.o.                             46.965** (19.821)     
## BuurtIJselbuurt Oost                              20.715 (13.380)      
## BuurtIJselbuurt West                              13.147 (14.760)      
## BuurtJacob Geelbuurt                              0.439 (35.231)       
## BuurtJacques Veldmanbuurt                         -4.501 (14.411)      
## BuurtJan Maijenbuurt                              9.378 (12.603)       
## BuurtJava-eiland                                  2.706 (17.084)       
## BuurtJohan Jongkindbuurt                          -3.148 (35.282)      
## BuurtJohannnes Vermeerbuurt                     63.062*** (14.138)     
## BuurtJohn Franklinbuurt                           7.940 (13.218)       
## BuurtJulianapark                                  2.471 (32.453)       
## BuurtK-buurt Midden                              -20.507 (54.839)      
## BuurtK-buurt Zuidoost                            -27.612 (30.855)      
## BuurtK-buurt Zuidwest                           -128.402* (76.745)     
## BuurtKadijken                                    33.967** (14.208)     
## BuurtKadoelen                                    -20.864 (19.761)      
## BuurtKalverdriehoek                             59.001*** (16.908)     
## BuurtKantershof                                  -12.207 (27.747)      
## BuurtKattenburg                                   6.718 (21.034)       
## BuurtKazernebuurt                                 26.966 (21.453)      
## BuurtKelbergen                                   -21.085 (45.075)      
## BuurtKNSM-eiland                                  22.409 (14.883)      
## BuurtKolenkitbuurt Noord                         -12.633 (19.292)      
## BuurtKolenkitbuurt Zuid                          -11.316 (15.550)      
## BuurtKoningin Wilhelminaplein                     17.071 (17.421)      
## BuurtKop Zeedijk                                57.007*** (15.070)     
## BuurtKop Zuidas                                  -13.790 (27.047)      
## BuurtKortenaerkwartier                           21.010* (12.324)      
## BuurtKortvoort                                    -7.313 (39.791)      
## BuurtKromme Mijdrechtbuurt                        9.217 (13.704)       
## BuurtL-buurt                                      -7.308 (18.376)      
## BuurtLaan van Spartaan                            17.335 (16.672)      
## BuurtLandelijk gebied Driemond                    20.437 (46.925)      
## BuurtLandlust Noord                               7.127 (12.959)       
## BuurtLandlust Zuid                                12.593 (11.073)      
## BuurtLangestraat e.o.                           78.452*** (13.039)     
## BuurtLastage                                    66.541*** (15.161)     
## BuurtLegmeerpleinbuurt                           26.057* (13.519)      
## BuurtLeidsebuurt Noordoost                      79.708*** (13.367)     
## BuurtLeidsebuurt Noordwest                      58.334*** (16.701)     
## BuurtLeidsebuurt Zuidoost                       56.561*** (16.549)     
## BuurtLeidsebuurt Zuidwest                       82.906*** (18.482)     
## BuurtLeidsegracht Noord                         75.210*** (15.624)     
## BuurtLeidsegracht Zuid                          87.187*** (15.923)     
## BuurtLeliegracht e.o.                           79.910*** (12.941)     
## BuurtLinnaeusparkbuurt                           23.250* (13.644)      
## BuurtLizzy Ansinghbuurt                          30.984** (12.536)     
## BuurtLoenermark                                   3.716 (28.501)       
## BuurtLootsbuurt                                  29.297** (12.326)     
## BuurtLouis Crispijnbuurt                         -12.846 (23.901)      
## BuurtLucas/Andreasziekenhuis e.o.                 8.827 (30.273)       
## BuurtMarathonbuurt Oost                           10.275 (14.847)      
## BuurtMarathonbuurt West                           10.361 (12.136)      
## BuurtMarcanti                                    34.746* (17.892)      
## BuurtMarine-Etablissement                         27.550 (22.065)      
## BuurtMarjoleinterrein                             15.026 (45.641)      
## BuurtMarkengouw Midden                            8.654 (23.136)       
## BuurtMarkengouw Noord                             0.352 (54.750)       
## BuurtMarkengouw Zuid                             -30.527 (77.427)      
## BuurtMarkthallen                                 -14.467 (27.017)      
## BuurtMarnixbuurt Midden                         52.587*** (18.366)     
## BuurtMarnixbuurt Noord                          52.212*** (13.867)     
## BuurtMarnixbuurt Zuid                             28.248 (18.388)      
## BuurtMedisch Centrum Slotervaart                  8.119 (76.304)       
## BuurtMeer en Oever                                -6.555 (22.580)      
## BuurtMercatorpark                                 -8.618 (28.595)      
## BuurtMiddelveldsche Akerpolder                   -10.055 (32.293)      
## BuurtMiddenmeer Noord                             2.136 (13.939)       
## BuurtMiddenmeer Zuid                              -3.485 (11.770)      
## BuurtMinervabuurt Midden                        59.670*** (17.911)     
## BuurtMinervabuurt Noord                         137.250*** (17.985)    
## BuurtMinervabuurt Zuid                           30.847* (16.988)      
## BuurtMolenwijk                                    7.879 (47.972)       
## BuurtMuseumplein                                  33.482 (30.397)      
## BuurtNDSM terrein                                 38.013 (24.132)      
## BuurtNes e.o.                                   61.427*** (15.508)     
## BuurtNieuw Sloten Noordoost                       19.301 (30.212)      
## BuurtNieuw Sloten Noordwest                       12.948 (20.509)      
## BuurtNieuw Sloten Zuidoost                        -6.390 (30.260)      
## BuurtNieuw Sloten Zuidwest                       -19.907 (26.601)      
## BuurtNieuwe Diep/Diemerpark                       27.710 (34.973)      
## BuurtNieuwe Kerk e.o.                           62.775*** (13.374)     
## BuurtNieuwe Meer                                 -17.787 (77.403)      
## BuurtNieuwe Oosterbegraafplaats                   -8.324 (54.505)      
## BuurtNieuwendammerdijk Oost                       5.505 (23.703)       
## BuurtNieuwendammerdijk Zuid                      -18.956 (35.639)      
## BuurtNieuwendammmerdijk West                      -0.222 (16.668)      
## BuurtNieuwendijk Noord                          53.961*** (17.194)     
## BuurtNieuwmarkt                                 95.116*** (14.936)     
## BuurtNintemanterrein                             -28.671 (54.595)      
## BuurtNoorder IJplas                            -560.382*** (78.785)    
## BuurtNoorderstrook Oost                           35.727 (77.329)      
## BuurtNoorderstrook West                          -20.494 (55.569)      
## BuurtNoordoever Sloterplas                       41.632** (16.688)     
## BuurtNoordoostkwadrant Indische buurt             8.097 (11.638)       
## BuurtNoordwestkwadrant Indische buurt Noord       10.170 (10.875)      
## BuurtNoordwestkwadrant Indische buurt Zuid        12.617 (11.035)      
## BuurtOlympisch Stadion e.o.                      -11.080 (21.226)      
## BuurtOokmeer                                      3.976 (35.074)       
## BuurtOostelijke Handelskade                      53.438** (23.312)     
## BuurtOostenburg                                   8.970 (13.605)       
## BuurtOosterdokseiland                           103.886*** (28.691)    
## BuurtOosterpark                                  36.641* (19.489)      
## BuurtOosterparkbuurt Noordwest                   21.189* (11.615)      
## BuurtOosterparkbuurt Zuidoost                    19.151* (11.582)      
## BuurtOosterparkbuurt Zuidwest                    22.854* (12.348)      
## BuurtOostoever Sloterplas                         12.604 (18.768)      
## BuurtOostpoort                                    9.940 (13.424)       
## BuurtOostzanerdijk                               -58.987* (30.654)     
## BuurtOrteliusbuurt Midden                         13.155 (12.993)      
## BuurtOrteliusbuurt Noord                          5.393 (14.167)       
## BuurtOrteliusbuurt Zuid                          21.303* (12.895)      
## BuurtOsdorp Midden Noord                          2.813 (28.524)       
## BuurtOsdorp Midden Zuid                          -13.031 (28.489)      
## BuurtOsdorp Zuidoost                              2.288 (18.314)       
## BuurtOsdorper Binnenpolder                       -28.462 (34.018)      
## BuurtOsdorper Bovenpolder                        -55.751 (46.204)      
## BuurtOsdorpplein e.o.                             -1.599 (24.680)      
## BuurtOude Kerk e.o.                             88.680*** (14.048)     
## BuurtOveramstel                                  -38.943 (78.582)      
## BuurtOverbraker Binnenpolder                      2.642 (39.450)       
## BuurtOverhoeks                                    -8.471 (28.812)      
## BuurtOvertoomse Veld Noord                        10.442 (14.895)      
## BuurtOvertoomse Veld Zuid                         -1.944 (15.673)      
## BuurtP.C. Hooftbuurt                            124.309*** (17.771)    
## BuurtPapaverweg e.o.                             34.358** (15.318)     
## BuurtParamariboplein e.o.                         13.248 (11.008)      
## BuurtPark de Meer                                 4.538 (23.197)       
## BuurtPark Haagseweg                              -33.029 (55.070)      
## BuurtParooldriehoek                              36.841** (18.254)     
## BuurtPasseerdersgrachtbuurt                     101.365*** (16.250)    
## BuurtPieter van der Doesbuurt                     4.515 (12.540)       
## BuurtPlan van Gool                                -3.698 (20.008)      
## BuurtPlanciusbuurt Noord                        154.868*** (28.469)    
## BuurtPlanciusbuurt Zuid                           34.560 (54.513)      
## BuurtPlantage                                   59.099*** (13.903)     
## BuurtPostjeskade e.o.                             16.998 (11.540)      
## BuurtPrinses Irenebuurt                           25.598 (19.521)      
## BuurtRAI                                          1.285 (28.473)       
## BuurtRansdorp                                     2.395 (34.602)       
## BuurtRapenburg                                  41.950*** (14.449)     
## BuurtRechte H-buurt                              -14.913 (31.052)      
## BuurtReguliersbuurt                             62.708*** (22.361)     
## BuurtReigersbos Midden                            0.656 (27.370)       
## BuurtReigersbos Noord                            -28.550 (27.384)      
## BuurtReigersbos Zuid                             -30.475 (39.331)      
## BuurtRembrandtpark Noord                          13.570 (18.052)      
## BuurtRembrandtpark Zuid                           1.266 (15.429)       
## BuurtRembrandtpleinbuurt                        55.087*** (16.197)     
## BuurtRI Oost terrein                              1.334 (17.115)       
## BuurtRieteiland Oost                              43.203 (35.955)      
## BuurtRietlanden                                   19.336 (15.679)      
## BuurtRijnbuurt Midden                             12.306 (13.507)      
## BuurtRijnbuurt Oost                              21.192* (12.649)      
## BuurtRijnbuurt West                               24.166 (19.644)      
## BuurtRobert Scottbuurt Oost                       -0.528 (14.182)      
## BuurtRobert Scottbuurt West                       -6.388 (13.206)      
## BuurtRode Kruisbuurt                             -42.921 (39.230)      
## BuurtSarphatiparkbuurt                          44.888*** (10.434)     
## BuurtSarphatistrook                             48.620*** (12.610)     
## BuurtScheepvaarthuisbuurt                       61.247*** (14.374)     
## BuurtScheldebuurt Midden                          13.446 (12.850)      
## BuurtScheldebuurt Oost                            13.921 (13.515)      
## BuurtScheldebuurt West                          32.905*** (12.108)     
## BuurtSchellingwoude Oost                         -20.788 (18.184)      
## BuurtSchellingwoude West                          15.337 (29.080)      
## BuurtSchinkelbuurt Noord                         22.049* (11.483)      
## BuurtSchinkelbuurt Zuid                           -1.373 (14.346)      
## BuurtSchipluidenbuurt                            -18.646 (39.209)      
## BuurtScience Park Noord                          -25.326 (18.232)      
## BuurtScience Park Zuid                           -15.565 (44.660)      
## BuurtSlotermeer Zuid                              4.268 (19.022)       
## BuurtSloterpark                                  -23.782 (32.128)      
## BuurtSloterweg e.o.                              -43.376 (29.676)      
## BuurtSpaarndammerbuurt Midden                     9.669 (16.993)       
## BuurtSpaarndammerbuurt Noordoost                  8.748 (13.413)       
## BuurtSpaarndammerbuurt Noordwest                  14.722 (18.096)      
## BuurtSpaarndammerbuurt Zuidoost                 40.224*** (15.425)     
## BuurtSpaarndammerbuurt Zuidwest                   14.183 (12.958)      
## BuurtSpiegelbuurt                               77.317*** (13.896)     
## BuurtSporenburg                                   16.444 (14.146)      
## BuurtSportpark Middenmeer Noord                   21.238 (44.697)      
## BuurtSportpark Middenmeer Zuid                   -33.635 (44.821)      
## BuurtSportpark Voorland                         153.078** (76.493)     
## BuurtSpuistraat Noord                           68.250*** (13.482)     
## BuurtSpuistraat Zuid                            77.142*** (15.916)     
## BuurtStaalmanbuurt                                11.387 (15.377)      
## BuurtStaatsliedenbuurt Noordoost                 29.695** (14.322)     
## BuurtStationsplein e.o.                           18.527 (76.586)      
## BuurtSteigereiland Noord                          -5.236 (18.046)      
## BuurtSteigereiland Zuid                          -16.147 (15.167)      
## BuurtSurinamepleinbuurt                           18.667 (14.021)      
## BuurtSwammerdambuurt                            36.237*** (11.890)     
## BuurtTeleport                                     13.365 (54.698)      
## BuurtTerrasdorp                                   -2.750 (18.774)      
## BuurtTransvaalbuurt Oost                          15.728 (11.435)      
## BuurtTransvaalbuurt West                          8.459 (11.885)       
## BuurtTrompbuurt                                  20.013* (11.199)      
## BuurtTuindorp Amstelstation                       11.242 (30.257)      
## BuurtTuindorp Frankendael                        -24.270 (20.284)      
## BuurtTuindorp Nieuwendam Oost                    -33.380* (17.267)     
## BuurtTuindorp Nieuwendam West                     -8.461 (23.350)      
## BuurtTuindorp Oostzaan Oost                      -22.166 (20.554)      
## BuurtTuindorp Oostzaan West                       1.991 (42.408)       
## BuurtTwiske Oost                                  -5.724 (56.857)      
## BuurtTwiske West                                 -16.365 (27.132)      
## BuurtUilenburg                                  50.065*** (15.528)     
## BuurtUtrechtsebuurt Zuid                        81.331*** (14.899)     
## BuurtValeriusbuurt Oost                         143.453*** (17.448)    
## BuurtValeriusbuurt West                          29.094** (14.588)     
## BuurtValkenburg                                 51.168*** (16.977)     
## BuurtVan Brakelkwartier                           24.482 (17.919)      
## BuurtVan der Helstpleinbuurt                    38.043*** (11.264)     
## BuurtVan der Kunbuurt                             17.229 (32.617)      
## BuurtVan der Pekbuurt                             18.104 (14.073)      
## BuurtVan Loonbuurt                              64.899*** (14.123)     
## BuurtVan Tuyllbuurt                              24.039** (12.128)     
## BuurtVelserpolder West                            -0.864 (18.161)      
## BuurtVeluwebuurt                                  27.693 (24.907)      
## BuurtVenserpolder Oost                            6.983 (18.951)       
## BuurtVliegenbos                                  30.749* (18.348)      
## BuurtVogelbuurt Noord                             0.660 (18.896)       
## BuurtVogelbuurt Zuid                              10.296 (12.825)      
## BuurtVogeltjeswei                                -41.887 (55.083)      
## BuurtVondelpark Oost                            222.054*** (39.257)    
## BuurtVondelpark West                             40.640* (23.101)      
## BuurtVondelparkbuurt Midden                     56.025*** (14.602)     
## BuurtVondelparkbuurt Oost                       49.935*** (14.228)     
## BuurtVondelparkbuurt West                       32.328*** (11.935)     
## BuurtVU-kwartier                                  17.262 (35.274)      
## BuurtWalvisbuurt                                  -2.932 (30.566)      
## BuurtWaterloopleinbuurt                         56.782*** (19.302)     
## BuurtWeesperbuurt                               48.865*** (12.069)     
## BuurtWeespertrekvaart                             27.345 (22.646)      
## BuurtWeesperzijde Midden/Zuid                     17.374 (11.621)      
## BuurtWerengouw Midden                             6.548 (18.799)       
## BuurtWerengouw Noord                              18.019 (54.737)      
## BuurtWerengouw Zuid                              -42.950* (24.471)     
## BuurtWestelijke eilanden                        44.903*** (12.544)     
## BuurtWesterdokseiland                             18.241 (15.565)      
## BuurtWestergasfabriek                             8.334 (21.213)       
## BuurtWesterstaatsman                             20.759* (11.094)      
## BuurtWestlandgrachtbuurt                          1.127 (10.936)       
## BuurtWeteringbuurt                              68.284*** (12.230)     
## BuurtWG-terrein                                 50.943*** (12.285)     
## BuurtWielingenbuurt                              35.890** (15.025)     
## BuurtWildeman                                    -30.603 (22.270)      
## BuurtWillemsparkbuurt Noord                     94.285*** (14.971)     
## BuurtWillibrordusbuurt                           26.405** (11.784)     
## BuurtWittenburg                                   19.359 (15.341)      
## BuurtWoon- en Groengebied Sloterdijk              16.308 (30.396)      
## BuurtZaagpoortbuurt                               24.086 (15.808)      
## BuurtZamenhofstraat e.o.                          19.933 (76.758)      
## BuurtZeeburgerdijk Oost                          -55.122 (54.562)      
## BuurtZeeburgereiland Noordoost                    27.169 (45.547)      
## BuurtZeeburgereiland Noordwest                    25.575 (39.263)      
## BuurtZeeburgereiland Zuidoost                    -11.476 (76.909)      
## BuurtZeeburgereiland Zuidwest                     5.873 (26.803)       
## BuurtZeeheldenbuurt                             34.190*** (12.658)     
## BuurtZorgvlied                                   -28.302 (45.551)      
## BuurtZuidas Noord                                 14.054 (39.275)      
## BuurtZuidas Zuid                                  -1.807 (21.834)      
## BuurtZuiderhof                                    34.334 (76.955)      
## BuurtZuiderkerkbuurt                            103.652*** (15.067)    
## BuurtZuidoostkwadrant Indische buurt              3.734 (13.171)       
## BuurtZuidwestkwadrant Indische buurt              6.375 (13.211)       
## BuurtZuidwestkwadrant Osdorp Noord               -16.907 (27.027)      
## BuurtZuidwestkwadrant Osdorp Zuid                 -2.947 (15.901)      
## BuurtZunderdorp                                  -24.548 (24.942)      
## minimum_nights                                    -0.030 (0.044)       
## convenienceshop.closey                           -12.880** (5.087)     
## mall.closey                                       5.973** (2.872)      
## museum.closey                                      2.385 (3.227)       
## plaza.closey                                       5.450 (5.895)       
## supermarkets.closey                               3.592 (11.073)       
## amenities.number                                 0.437*** (0.069)      
## lagRevenue                                      -0.00002 (0.00003)     
## metro.dist                                        -5.364 (5.335)       
## Constant                                          39.299 (26.996)      
## -----------------------------------------------------------------------
## Observations                                          19,952           
## R2                                                     0.427           
## Adjusted R2                                            0.412           
## Residual Std. Error                             75.861 (df = 19453)    
## F Statistic                                 29.054*** (df = 498; 19453)
## =======================================================================
## Note:                                       *p<0.1; **p<0.05; ***p<0.01

MAPE (Mean Absolute Percentage Error)

By comparing the predicted price and actual price of our testing set, we summarized the errors in the table and map down below. Our model has a mean absolute percentage error of 29%, and it seems to be doing a better job in the central area than peripheral areas. As we will show in the following section, our model’s general predictibility is higher in central areas, given the abundance of existing data concentrating in that area.

annualrev_predict_test <- annualrevenue.test %>%
  mutate(Prediction = predict(reg.annualrevenue, newdata = annualrevenue.test)) %>%
  mutate(Prediction = ifelse(Prediction > 0, Prediction, mean(annualrevenue.training$price)))%>%
  filter(price!=0)%>%
  drop_na(Prediction)%>%
  mutate(AE = abs(Prediction-price),
         APE = AE/Prediction)

test_result <- data.frame(MAE = c(mean(annualrev_predict_test$AE, na.rm=T)),
              MAPE = c(scales::percent(mean(annualrev_predict_test$APE, na.rm=T)))) 

test_result %>%
  kable(caption = "Figure 8. Mean absolute error and MAPE for a single test set")%>%
  kable_styling("striped", full_width = F)

Figure 8. Mean absolute error and MAPE for a single test set

MAE	MAPE
42.91289	30%

palette5 <- c("#ffffd4", "#fed98e", "#fe9929", "#d95f0e", "#993404")

annualrev_predict_test%>%
  group_by(Buurt) %>%
  dplyr::summarize(mean.APE = mean(APE, na.rm = T)) %>%
  ungroup() %>% 
  left_join(neighbor2,by = "Buurt") %>%
    st_sf() %>%
    ggplot() + 
      geom_sf(aes(fill = mean.APE),colour = 'transparent') +
      scale_fill_gradient(low = palette5[1], high = palette5[5],
                          name = "MAPE") +
      labs(title = "Average Test Set MAPE"
           #,subtitle = "2019"
           ) +
      mapTheme()

Cross Validation: LOGO (Leave One Group Out)

LOGO (Leave One Group Out) is a type of cross validation that is similar to K-fold but takes spatial aspects into consideration – taking one neighborhood out at a time for cross validating. Since the airbnb location is critical in this project, we think LOGO would be a more suitable way for testing out our model. Here we use 481 ‘buurt (buurten)’ (neighborhood in dutch) defined by the city.

reg.vars <- c("beds", "bedrooms", "bathrooms", "accommodates",
                              "pool", "parking", "kitchen", "AC", "fireplace",
                              "room_type",  "bed_type",
                              "amenities.number","lagRevenue", "convenienceshop.close", "mall.close",
                              "museum.close",
                              "plaza.close", "supermarkets.close")

airbnb.location <- details.sf %>% dplyr::select(id)

annualrev_predict_test <- annualrev_predict_test %>% left_join(airbnb.location, by='id')
colnames(annualrev_predict_test)[1] =  "identifier"

reg.spatialCV <- crossValidate(
  dataset = annualrev_predict_test,
  id = "Buurt",
  dependentVariable = "price",
  indVariables = reg.vars) %>%
    dplyr::select(identifier = Buurt, price, Prediction, geometry)

reg.spatialCV <-
  reg.spatialCV %>%
  mutate(Error = Prediction - price)

error_by_reg_and_fold <-
  reg.spatialCV  %>%
  group_by(identifier) %>%
  dplyr::summarize(Mean_error = mean(Error, na.rm=T),
                   MAE = mean(abs(Mean_error), na.rm=T),
                   SD_MAE = mean(abs(Mean_error), na.rm=T)) %>%
  ungroup()

st_drop_geometry(error_by_reg_and_fold) %>%
    dplyr::summarize(Mean_MAE = round(mean(MAE), 2),
              SD_MAE = round(sd(MAE), 2)) %>%
  kable() %>%
    kable_styling("striped", full_width = F) 

Mean_MAE	SD_MAE
24.58	25.45

From the LOGO-CV result, our mean absolute error is 25.1 euro, with a 23.77 euro standard deviation. This means that on average our prediction could be 25.1 euro off, higher or lower. Given our predictability being ~42%, this is an expected and acceptable error to be observed.

Since earlier we observed a clustering in central areas, we are curious to see how our model performs in areas with more existing Airbnb vs. less. Hence we split neighborhoods into those with more than 50 Airbnb vs. less than 50, checked the respective mean absolute error for the high vs. low count groups, then plotted the mean error in each neighborhood.

#Counting per neighborhood
neighbourhoodCount <- annualrevenue %>%
  dplyr::count(Buurt)

#Joining to new spatial db
neighbourhoodHL <- neighbor2 %>%
  left_join(st_drop_geometry(neighbourhoodCount), by = "Buurt")%>%
  mutate(Airbnb1k = ifelse(n < 50, "Low", "High"))%>%
  st_transform(st_crs(error_by_reg_and_fold))

#spatial join of the essential comparison data
st_join(error_by_reg_and_fold, neighbourhoodHL) %>% #fill in data frame with error column here
  group_by(Airbnb1k) %>%
  dplyr::summarize(mean.MAE = scales::percent(mean(MAE, na.rm = T))) %>% #fill in error column here
  st_drop_geometry() %>%
  spread(Airbnb1k, mean.MAE) %>%
  kable(caption = "<strong>Test set MAE within Neighborhoods Above or Below 50 Listings </strong>")%>%
  kable_styling()

Test set MAE within Neighborhoods Above or Below 50 Listings

High	Low
1 585%	2 880%

The mean absolute error in area with higher count of Airbnb is significantly lower than the area with lower count, according to the data. The map below has lower-count areas in grey shade. By comparing shaded (lower count) to unshaded (higher count) areas, we can see that the shaded areas have more variation while the unshaded areas have mean errors closer to zero – suggesting a better performance in higher count area.

colnames(error_by_reg_and_fold)[1] = "Buurt"

new <-
  neighbor2 %>% dplyr::select(Buurt, geometry) %>%
  left_join(st_drop_geometry(error_by_reg_and_fold), by = 'Buurt')
  

Low_Count_Boundary <-
  st_union(neighbourhoodHL %>% filter(Airbnb1k == 'Low'))

 mycol <- rgb(220,220,220, max = 255, alpha = 90, names = "blue50")

ggplot() +
  geom_sf(data=new, aes(fill=Mean_error), color=NA) +
  scale_fill_gradient(low = '#91bfdb', high = '#fc8d59', na.value = 'white', name = "Mean Error") +
  labs(title = "Mean Error (LOGO-CV)", subtitle = "Amsterdam, NL") + 
  geom_sf(data=Low_Count_Boundary, color="black", fill = mycol, size=0.7) + mapTheme() 

Conclusion

In conclusion, our model gives a decent prediction for airbnb price in Amsterdam, although for it to be launched for our users, new hosts in Amsterdam, we would need more upgrades and feature engineering. From the results of cross validation, our model is better predicting price in areas with higher concentration of existing Airbnb in comparison to lower density areas, given the fact that more existing data is provided. We hope to improve this model / app and launch it for new hosts to utilize and better price their units in the near future. We will come insightful comments and feedback from Amsterdam locals.

In addition, as this report demonstrate, many features incorporated in this model are linked specifically to Amsterdam, such as property type (given Amsterdam’s boat house and other interesting architectural types) and specific amenities (such as fireplaces). At this stage, we are refraining from attempting to apply this model to other cities around the world. At the end of the day, each city (especially cities famous for tourism) have their unique selling points that attract tourists, and that is the reason why tourism and Airbnb exist in the first place.

While our model is primarily built for new hosts, some of the analysis included in this report would also be helpful for policy makers as risk assessment. Although it is reasonable to try to develop all areas in the city, simply banning short-term rental in central areas may be detrimental to tourism yet not benefiting the peripheral areas of the city. By transforming this price-prediction model into a risk-prediction model, policy-makers will have more insights on which areas are potentially exploiting short-term rental. Policy-makers and planners can also work together to install amenities that are favorable by tourists, such as public transit and convenience shops, to direct flow of tourists to certain areas. In sum, information in this report could be helpful to policy makers as well.