relative weight analysis<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />How can I calculate relative weights to see how independent variables contribute to a multiple regression when there is multicollinearity between the independent variables ?
Thanks,
Marcus
Re: relative weight analysisUse principal components regression. First, run a principal components on the standardized explanatory variables. Then, run a least-squares regression using the principal components to predict your y-variable. As the principal components were caclulated from standardaized varibles, the regression coeffients will be the relative weights.
