Linear regression is a statistical method used to analyze the relationship between dependent and independent variables. Parameter estimation is generally obtained through the Ordinary Least Square (OLS) method, which produces unbiased and efficient estimates. However, the presence of multicollinearity and autocorrelation can render OLS estimates suboptimal. The Ridge Regression method combined with Prais Winsten correction can to produce more accurate parameter estimates than OLS. Unlike the subjective approach in determining the bias constant () through Ridge Trace, this study determines the optimal value by minimizing the Mean Square Error (MSE). The results show that the Ridge Regression model with Prais Winsten autocorrelation correction has an Adjusted R² of 78% and a Root Mean Square Error (RMSE) of Rp364,389. Four independent variables are found to have a significant effect on the exchange rate of the Indonesian Rupiah against the United States Dollar (USD), namely money supply, interest rate, exports, and imports.
Copyrights © 2026