A basic script in R for how to pull data from ImmobilienScout24's Rest Price History API

ImmobilienScout24 is Germany's largest real estate website (akin to Zillow). Their API documentation hasn't been updated in ages and does not have formal support for those who wish to access it via R. There are some examples in Python, but since I am not familiar with that and I needed to access their data, I went ahead and figured out how to obtain data using R.

Geocode addresses ahead of time

It's complicated to geocode addresses using ImmobilienScout's API -- so I used Google and/or the Data Science Kit (which is essentially the Open Street Map) to geocode my list of addresses (for this project that was about 5,000 unique addresses).

Fun w/ ImmoScout

The final script I wrote in R to scrape the Immo data is called: ImmoScout_API_PriceHistory_KyleOtt.R

First, you have to go to the ImmoScout “Playground”

https://playground.immobilienscout24.de/rest/playground

Here, you enter our own consumer key and secret and select “live”: consumerKey <- 'SECRET KEY' consumerSecret <- 'SUPER SECRET KEY'

Then scroll down to see oAuth-Frame* and click request token. The tokens should generate in the empty boxes. Next, click “authorize” below and you should be taken to a log in page. You can use any ImmoScout log-in (it's easy to generate a log-in). Authorize and it will take you back to the playground. Now, click the last orange box called “access token”, this will give you the access token and token secret which is needed to preform the O-auth in R Studio (you just copy and paste these two strings). You can test API URLS in the URL box on the playground now and if you enter a correct URL, it will give you actual output. Always select JSON and GET as options (no idea why, but based on the forums I've read, this is how you should go about things).

ImmoScout’s Reference API

Use this API to get current housing price data. Their “codebook” of how to use the Reference API and what kind of output you can get is here:

http://api.immobilienscout24.de/our-apis/valuation/reference-price-api.html

The documentation is sparse and not very helpful and they have incomplete information for the kinds of URLS you should request Important things to know:

• The housing data is based on rental apartments
• The assumed living area is 100 square meters
• The price is the price per square meter in a given neighborhood
• ImmoScout does not specify, but the historical data should be updated to about 3 months prior to
• ImmoScout now only lists price information by one decimal (previously it was two)

With this, I ran a R script in this repo and it worked! It’s a simple script that logs into our API key and gets access and authorization to get “live” data from ImmoScout.