A Comparison of WK4 and MSE for Regression Model Fitting

Abstract

Wasfi Kahwachi4 (WK4) measurement is considered as one of the powerful statistical tools to testing goodness of fit of regression model when compared with the MSE (Mean Squared Error) measure. In this paper, the main goal is to detect the best measure for model fitting by using a comparison between MSE and WK4 measure. This new measure (WK4) applied on five samples to compare its value with MSE value which obtained from the same samples to judge on the new criterion for fitting the models. The results showed that the WK4 measure was better than the MSE measure (smaller) for fitting regression model in all samples. This gives the conclusion that Wk4 can be used besides MSE for testing goodness of fit in regression analysis.