Multicollinearity and autocorrelation are common problems in regression analysis. This study tested a new regression model called GLS-Ridge on simulated data to see how well it handles both issues at once. The GLS-Ridge model outperformed traditional methods like least squares, ridge regression, and LASSO in predicting outcomes accurately. This means that the GLS-Ridge model is better at making predictions when there are multicollinearity and autocorrelation in the data.