The purpose of this project is to develop a model that is able to predict property values in Los Angeles, California, using the Zillow dataset. We intend to discover the key features that most accurately predict property value.
In addition, we have provided these deliverables: 1. Map showing variations in tax rates by county. 2. Acquire.py, Prep.py, Wrangle.py, and Model.py files to be able to recreate our efforts. 3. Presentation with the results of our findings.
Generated with ryans_codeup_data_science_mvp
Modified from datasciencemvp
├── README.md <- The top-level README for developers using this project.
│
├── data <- All of the data for the project
│ ├── modeling <- The prepared, processed and split datasets for modeling.
│ ├── prepared <- The prepared datasets for exploration
│ └── raw <- The original, immutable data
│
├── main.py <- The main python script that calls all src scripts
│
├── mvp.ipynb <- The main notebook for the project
│
├── src <- The source code for use in this project
│ ├── __init__.py <- Makes src a Python module
│ ├── acquire.py <- The script to download or generate data and store it in
│ │ data/raw/
│ ├── explore.py <- The script for creating any visuals that need to be stored
│ │ in visuals/generated_graphics/
│ ├── model.py <- The script for preprocessing, modeling, and interpreting
│ └── prepare.py <- The script for preparing the raw data and storing it in
│ data/prepared/
│
└── visuals <- All project visuals
├── external_visuals <- Visuals brought from outside the project
├── generated_graphics <- Visuals generated from the project
└── presentation <- A copy of your presentation
Initial Questions:
- Does number of bedrooms matter to the property value?
- Does location within the state or county affect overal property value?
- Is total square footage independent of property value?
- Does a combined bedroom/bathroom square footage feature have better correlation with the target variable than the separate features?
-
H0: Properties with 2 or more bedrooms do not have higher value than average property
-
Ha: Properties with 2 or more bedrooms do have higher value than average property
-
H0: Properties with 2 or more bedrooms do not have more square footage than average property
-
Ha: Properties with 2 or more bedrooms do not have more square footage than average property
| Feature | Definition |
|---|---|
| bathroomcnt | Number of bathrooms in property (includes half bathrooms) |
| bedroomcnt | Number of bedrooms in property |
| calculatedbathnbr | Number of both bedrooms and bathrooms in property |
| calculatedfinishedsquarefeet | Total Square Footage of the property |
| fullbathcnt | Number of full bathrooms in property (excludes half bathrooms) |
| Target | Definition |
|---|---|
| taxvaluedollarcnt | Value of the property |
-
We found that removing outliers for actual home values column had overfit the model.
-
The model performed very well on the majority of the population home values.
-
Based on our heat map of the local Los Angeles counties, LA county is paying the most in taxes compared to its neighbors
-
In order to improve on our model, we would need more listings of the homes that were in the outlier range to adjust for higher valued homes.
-
The model did perform better than the baseline significantly, but there is room for improvement with more data.