westonkl/housing-price-prediction

Predicted housing prices with several different regression models and xgboost.

Housing Price Prediction: Project Overview

• Data preprocessing: took log of skewed numerics, created dummies for categoricals, and replaced null values with the mean of their respective column
• Predicted housing prices with ridge, lasso, and elastic regression models
• Found the most impactful coefficients(features) on predicting a house's price with lasso and elastic
• Used xgboost to get a better score

Resources Used:

Python Version: 3.6 Packages: pandas, numpy, sklearn, matplotlib, scipy, xgboost

Data Preprocessing:

• took log of skewed numerics
• created dummies for categoricals
• replaced null values with the mean of their respective column

Model Building and EDA:

Finding alpha for ridge regression:

Finding the most impactful coefficients with lasso regression:

Plotting residuals:

Model Evaluation:

I tested 4 different modesls and evaluated them with root mean square error (RMSE).

• Ridge Regression : 0.12733734668670765
• Lasso Regression : 0.1225673588504812
• Elastic Regression : 0.1237462980365083
• Xgboost : 0.12494475766362181