The work performed here is based on data from the Kaggle competition below:
https://www.kaggle.com/competitions/house-prices-advanced-regression-techniques
From the perspective of an online real estate market company, there are two challenges:
- Accurately assess home values by neighborhood
- Determine importance of various attributes in determining that value
Developing a data-driven / model-based approach can potentially:
- Provide more precision in valuation methodology
- Assess what factors drive the overall value of a home
Zillow
Zillow is an online real estate marketplace company.
Zillow is famous for using state of the art statistical and machine learning models to produce Zestimates (their trademarked valuation model).
To calculate a Zestimate, Zillow uses a sophisticated neural network-based model that incorporates data from the following sources:
- Market trends
- On-market data
- Off-market data
- Home characteristics
- County and tax assessor records
- Direct feeds from listing services and brokerages
https://www.zillow.com/research/zestimate-forecast-methodology/
Opendoor
Opendoor is an online real estate marketplace company. According to an article written by the company, there are * primary factors that influence home value:
- Location
- Interest rates
- The local market
- Age and condition
- Economic indicators
- Neighborhood comps
- Upgrades and updates
- Home size and usable space
https://www.opendoor.com/w/blog/factors-that-influence-home-value
X: 79 explanatory variables describing aspects of residential homes in Ames, Iowa y: Home price
Develop a model to predict home price based on a collection of attributes about the home and surrounding neighborhood.