Project Zillow

Predict the tax assessed property value of Zillow Single Family Residential properties with transaction date in 2017

Project Description

As the most-visited real estate website in the United States, Zillow and its affiliates offer customers an on-demand experience for selling, buying, renting and financing with transparency and nearly seamless end-to-end service. I have decided to look into the different elements that determine tax assessed property value.

Project Goal

• Discover drivers of property value
• Use drivers to develop a machine learning model to predict property value
• This information could be used to further our understanding of how Single Family properties are tax assessed for their property value

Initial Thoughts

My initial hypothesis is that drivers of tax assessed property value will be the elements like number of rooms, square feet, and location.

The Plan

• Acquire data from Codeup MySQL DB
• Prepare data
  • Create Engineered columns from existing data
• Explore data in search of drivers of property value
  • Answer the following initial questions
    • Is there a correlation between area and property value?
    • Is there a correlation between age and property value?
    • Is there a correlation between the room count and property value?
    • Is there a difference in average property value between counties?
• Develop a Model to predict property value
  • Use drivers identified in explore to help build predictive models of different types
  • Evaluate models on train and validate data
  • Select the best model based on $RMSE$ and $R^2$
  • Evaluate the best model on test data
• Draw conclusions

Data Dictionary

Original	Feature	Type	Definition
yearbuilt	year	Year	The year the principal residence was built
bedroomcnt	beds	Numeric	Number of bedrooms in home
bathroomcnt	baths	Numeric	Number of bathrooms in home including fractional
roomcnt	roomcnt	Numeric	Total number of rooms in the property
calculatedfinishedsquarefeet	area	SqFt	Calculated total finished living area
taxvaluedollarcnt (target)	prop_value	USD	The total tax assessed value of the parcel/home
fips	county	County	Federal Information Processing Standard (these 3 in CA)
latitude	latitude	Numeric	Latitude coordinates of property
longitude	longitude	Numeric	Longitude coordinates of property
Additional Features		Numeric	Encoded categorical variables
	age	Year	How many years from 2017 since it was built

FIPS County Codes:

• 06037 = LA County, CA
• 06059 = Orange County, CA
• 06111 = Ventura County, CA

Steps to Reproduce

1. Clone this repo
2. If you have access to the Codeup MySQL DB:
   • Save env.py in the repo w/ user, password, and host variables
   • Run notebook
3. If you don't have access:
   • Request access from Codeup
   • Do step 2

Conclusions

Takeaways and Key Findings

• The younger the property the better for property value
• The bigger the living area the bigger the property value
• Location matters for property value
• Model still needs improvement

Recommendations and Next Steps

• It would nice to have the data to check if the included appliances or the type of heating services (gas or electric) of the property would affect property value
• More time is needed to work on features to better improve the model
  • latitude and longitude could hopefully give insights into cities and neighborhoods with higher or lower property values
  • pools and garages could also be looked into