3

I am a student with the Singapore Management University's postgraduate School of Economics. I am writing to seek your assistance in obtaining certain data on San Francisco’s housing market.

For our Econometrics project - we are looking into running a regression model to predict the housing prices in San Francisco. To do so, we would require datasets of houses in San Francisco city.

I am wondering if it is possible for you to share with us the raw data collected to arrive at these reports. Or others, to point us towards the person and organize we would be able to obtain it from.

We would require 10 years worth of monthly data:

  • Housing purchase prices in San Francisco
  • Per sq ft area of houses
  • Types of houses (house, apartment)
  • Number of bedrooms
  • Extra amenities (pool, tennis court)
  • Quality of neighborhood
  • Age of house
  • Size of yard
  • Location (from city center)
  • housing.datasf.org looked promising, but isn't nearly detailed enough. By the way, if you can predict the housing prices, then the data to do so won't be open, since people are making money out of being the only ones with it. That means you'll likely have to scrape a real estate website. – philshem Nov 5 '16 at 7:28
  • I was just going to say Zillow too! Data from Zillow is also available on Quandl. See, for example: quandl.com/data/… – user37317 Nov 10 '16 at 18:58
  • @philshem, You can predict everything, but does it match reality? – Pacerier Oct 28 '17 at 20:24
3

Zillow is an online real estate database company and offers research data which is aggregated, and you may find meets some of your needs.

http://www.zillow.com/research/data

All data accessed and downloaded from this page is free for public use by consumers, media, analysts, academics etc., consistent with our published Terms of Use. Proper and clear attribution of all data to Zillow is required.

Additionally, Zillow writes:

For other data requests or inquiries for Zillow Real Estate Research, contact us here.


Interesting graphic:

enter image description here

2

Some kind soul scraped the newspaper listings and put the results on github here.

However, several of the things you ask are much more involved, require analysis and subjective judgment to furninsh, and won't be found in standard real estate datasets. For example, Quality of neighborhood can mean any number of things, even for hedonic modeling purposes. Similarly, location from city center requires some judgment as to what you think that center is, let alone running the calculation. While a straightforward GIS process, it still requires some judgment depending on whether your model is mainly focused on the employment center or the retail center (further south west).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.