Real-Estate-Tycoon

Project Context

Our client is a large Real Estate Investment Trust (REIT).

They invest in houses, apartments, and condos within a small county in New York state. As part of their business, they try to predict the fair transaction price of a property before it's sold. They do so to calibrate their internal pricing models and keep a pulse on the market.

Current Solution

The REIT currently uses a third-party appraisal service. Appraisers are professionals who visit a property and estimate a fair price using their own expertise.

Unfortunately, the skill levels of individual appraisers vary greatly. During a trial run, the REIT compared appraiser estimates to actual transaction prices. The REIT found that the estimates given by inexperienced appraisers were off by $70,000, on average!

Our Role

The REIT has hired us to find a data-driven approach to valuing properties.

They currently have an untapped dataset of transaction prices for previous properties on the market. Our task is to build a real-estate pricing model using that dataset. If we can build a model to predict transaction prices with an average error of under $70,000, then our client can replace inexperienced appraisers with our model.

Problem Specifics

It's always helpful to scope the problem before starting.

Deliverable: Trained model file

Machine learning task: Regression

Target variable: Transaction Price

Win condition: Avg. prediction error < $70,000

Now that you have the context and problem specifics, let's dive right in!

Download (Dataset+Data Dictionary)

https://drive.google.com/open?id=1rl7CrlSzhW9nR7VHhx8az_0SbIBBrxkB

cleaned_df.csv

https://drive.google.com/open?id=1KC8V4265gIkiHZmKrgol79rrmIwvOkWP

analytical_base_table.csv

https://drive.google.com/open?id=17ITZtZJMku17vqoSFzfjo2vh38forfyL

final_model.pkl

https://drive.google.com/open?id=1sffYOwOUOw69GgD5y8yoOZyLqAtHO8Wy