Abstract :
The starting point is the fact that I’m hearing more and more about Austin housing market :
- Big tech companies are massively investing in Texas : Tesla, Google, Space X …
- All those companies are bringing highly paid worker who modify Austin housing market
- Airbnb is a known problem regarding housing since it’s an investment that contributes to very high rent
Second point to consider, I have been to Austin several times and I love this city, It is great for outdoors and that will be part of my opinion.
I don’t think the outdoor surroundings of Austin are prepared to welcome a massive 2 or 3 millions people, it;s going to become an overcrowded hell very fast.
With this in mind, and knowing that Airbnb is nowadays one of the biggest most popular online marketplaces for accommodations, I looked into the data to:
- On one hand evaluate the impact Airbnb has on housing in Austin;
- On the other hand investigate about who, when and how rents out in Austin.
You can find my script on Github : Airbnb analysus github
Data from : http://insideairbnb.com/index.html
Files used : 2 csv files, listing and listing detailed
| Feature | Cleaning action | Method |
|---|---|---|
| neighbourhood_group | Fill in the names corresponding to the Zip codes | Mapping with a dictionary and replace method |
| license | drop the column | pd.drop |
| Price | Analysis of the price feature who didn't include fees and cleaning | Creation of a cleaning fee proportional to number of bedrooms in the Airbnb |
| Last_review | Parse dates | pd.to_datetime |
Distribution :
We can already notice that despite Airbnb being originally created to share rooms, in Austin more than 80% of the listings are actually about entire homes or apartments or even hotel rooms, while private and shared rooms make less than 20%. This is already an alarming piece of information.
We already knew that Airbnb was used as an investment but the proportions are staggering.
% distribution of listings :
South central and Central Austin are really the sparkling area of Austin, close to UT and very practical for tourists.
East Austin is obviously an areas where investors try to buy because it’s relatively cheap compared to central.
Travis county represent the part of the population with large lots or home and using this to make profit.
Listing per users :
Here we focus our data analysis on how much of real estate is concentrated for investment rather than resident housing.
The results are here, in Downtown Austin almost every host has on average 2 listings, making actual apartment for resident harder to find.
Price distribution :
Since they are still slightly skewed, I normalize prices with a logaritmic transformation.
Let’s see if price has different distribution depending on the language. To do this, we will do a graphical evaluation and then we will apply a statistical test (T-test for independent samples)
Log transformation is mainly used if you want to pursue a predictive model, they tend to work better with those distribution.
You’re so awesome! I don’t believe I’ve read something
like this before. So wonderful to find another person with a few unique thoughts on this issue.
Really.. thanks for starting this up. This website is one thing that is needed on the web, someone with a little originality!
Hi Tadalafil !
Thank you so much for your feedback, I appreciate everything you said toward this article !
Howdy! I know this is kind of off topic but I was wondering which blog platform are you using for
this site? I’m getting sick and tired of WordPress because I’ve had problems with hackers and
I’m looking at alternatives for another platform.
I would be awesome if you could point me in the direction of a good platform.