The investigation of factors that could affect systolic blood pressure with ridge regression analysis and multicollinearity problem

The purpose of this study is to investigate the effectiveness of biased estimation technique Ridge regression over Least Squares technique in predicting systolic blood pressure. For this purpose, initially the problem of multicollinearity was examined then, the relative predictive validity of Least Squares and Ridge regression techniques was compared in predicting linear relation between systolic blood pressure and independent variables. In this study, by virtue of high degree of multicollinearity problem between independent variables, it was determined that Ridge regression technique would provide more stabilized coefficents and less standard error than would the Least Squares technique.Key words: Regression Analysis, Multicollinearity Problem, Ridge Regression