Abstract:
The maximum mean squares error will result when the least squares method is used to determine the regression coefficients of ill condition of the independent variables. In fact, the approximate value of parameter β, is not good enough in quality. This research shows how to evaluate the regression coefficients by the method of Ridge Regression. This method has presumably more effectiveness in minimizing the mean squares error than the method of the least squares. The study, in addition,-shows the evaluation of regression coefficients by the method of Ridge Regression. This method has presumably more effectiveness in minimizing the mean squares error than the method of the least squares. The study, in addition, shows the evaluation of regression coefficients between the method of least squares and Ridge Regression. The result of this research by Ridge Regression method gives the biased estimator of parameter β and minimizes the mean squares error when compared to the least squares method. The results of these two methods have equal pros and cons when using them to examine the estimated value of both methods with the observed variable. Given the estimate values of each method, the value of the above method is not different from the observed variable of the dependent variable. The Ridge Regression method is not suitable for general data like the least squares method. The Ridge Regression method is very difficult, complicated and time consuming for the purpose of statistic value determination. On the other hand, if the data concerning the determination of the independent variables approach to non-orthogonal, the Ridge Regression is the method for the estimation of the regression coefficient. This method is better than the least squares method because it gives the minimum mean squares error, compared to the least squares method.