What question did this study set out to answer?

This research investigates how regularization estimators perform under severe multicollinearity conditions in housing price prediction.

Evaluating Regularization Estimators Under Severe Multicollinearity: A Simulation and Empirical Study on Housing Prices

View Full Paper

Ask AI

Bookmark

Authors

Simulation study evaluates housing price prediction with regularization methods under multicollinearity, indicating practical modeling strategies.

This research investigates how regularization estimators perform under severe multicollinearity conditions in housing price prediction.
Conducted simulation experiments with sample sizes from 100 to 2000 and 5 to 70 independent variables under controlled correlation structures.
Applied regularization methods like ridge regression, LASSO, and elastic net, evaluating them on the California Housing Dataset.
Calculated mean squared prediction error (MSPE), root mean squared error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) on a hold-out test set.
LASSO generally demonstrates the lowest variance estimates under moderate multicollinearity.
Elastic net becomes more competitive as multicollinearity and the number of independent variables increase.
Elastic net reduces MSPE by approximately 95.5% compared to ordinary least squares (OLS) in the California Housing application.

Ekiz et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Simulation study evaluates housing price prediction with regularization methods under multicollinearity, indicating practical modeling strategies.