https://github.com/grantgasser/kaggle-home-prices
Done with Matt Suski as part of the Kaggle Competition
https://github.com/grantgasser/kaggle-home-prices
best-subsets housing-prices kaggle lasso regression
Last synced: 7 months ago
JSON representation
Done with Matt Suski as part of the Kaggle Competition
- Host: GitHub
- URL: https://github.com/grantgasser/kaggle-home-prices
- Owner: grantgasser
- Created: 2018-11-19T21:43:44.000Z (almost 7 years ago)
- Default Branch: master
- Last Pushed: 2018-11-27T19:01:28.000Z (almost 7 years ago)
- Last Synced: 2025-02-08T18:14:43.440Z (8 months ago)
- Topics: best-subsets, housing-prices, kaggle, lasso, regression
- Language: R
- Size: 271 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Predicting Home Prices (Kaggle Competition)
## Prelim Report:
### Main
This project was done jointly with [Matt Suski](https://www.linkedin.com/in/matt-suski/) for [Dr. Patrick's](https://www.baylor.edu/statistics/index.php?id=941853) Regression 3386 course.
### Methodology
With a dataset of 79 features, we used a combination of regsubsets and Lasso to perform model selection. Model with variables selected by Lasso (glmnet in R) performed better with a RMSE = .16044. We examined outliers & influential points. We also tested the following assumptions of regression models:
1. Normality of residuals
2. Independence of residuals
3. Constant variance of residuals