Skip to content

Latest commit

 

History

History
20 lines (11 loc) · 1.12 KB

README.md

File metadata and controls

20 lines (11 loc) · 1.12 KB

This is Python script that works with 2015 U.S. Census CSV data and recent Zillow data dump on median home prices. The purpose of this project is to understand the relationship between median income and median home sales in US counties and cities. It's difficult to determine where in this country your family's dollar goes the furthest, and this project attempts answer that.

He is the source data used in this project so far: -http://files.zillowstatic.com/research/public/County/County_Zhvi_Summary_AllHomes.csv -https://www.census.gov/did/www/saipe/downloads/estmod15/index.html

To evaluate where a family's dollar goes first, we use Zillow's home value index to derive a Wellness Index for each city and state.

Wellness Index = Zillow Home Value Index / median income

The output like this:

[('UT_Utah', 4.094598243356999), ('HI_Hawaii', 4.640409319123904), ('NV_Carson City', 5.61948314367755), ('DC_District of Columbia', 7.6441222731313685), ('NY_New York', 20.94564591727654)]

A High Wellness index value indicates the relative expense to live in a region.

Other factors / API calls to add later may include:

  • Walk score