https://communities.sas.com/t5/Statistical-Procedures/Regression-analysis-suggestion/m-p/71268#M3449 The only time statements like "var x1 has 20% var x2 has 12%" have meaning is when the variables are not correlated with one another.<BR /> <BR /> As soon as you have correlations, statements of the sort "var x1 has 20% var x2 has 12%" are meaningless.<BR /> <BR /> I suggest you adopt a different mindset. You can still determine which <I>combinations </I>of variables are important in the prediction equation; you cannot determine which individual variable is causing the impact via statistics alone, nor can you quantify the percent impact of an individual variable.<BR /> <BR /> One technique that allows you to determine the combinations of variables that are important in the predicting a response(s) is Partial Least Squares regression. Mon, 13 Sep 2010 13:08:24 GMT Paige 2010-09-13T13:08:24Z https://communities.sas.com/t5/Statistical-Procedures/Regression-analysis-suggestion/m-p/71267#M3448 Hi all,<BR /> <BR /> I am currently running regression analysis to find out Impact of Each indep variable on dep variable<BR /> <BR /> say,<BR /> <BR /> y = b1x1 + b2x2 + b3x3 + b4x4 ....<BR /> <BR /> The output that I seek is var x1 accounts for 20% , var x2 accounts for 30% impact. to calculate this I am using standardized coefficients. <BR /> <BR /> However, I see that there are a lot of variables that have <B>correlation amongst them.</B><BR /> <BR /> I know factor analysis and principle analysis are methods of dealing with the problem of correlation .. however, in the end I want to see the impact in terms of percentage .. Like var x1 has 20% var x2 has 12% and so on ..<BR /> <BR /> <B> what correlation would be considered as high .. like above 0.7 or above 0.8 or some number ... </B><BR /> <BR /> It would be helpful if you guys could suggest me some method to this .. <BR /> <BR /> Thank you,<BR /> <BR /> Rockerd. Sat, 11 Sep 2010 08:48:15 GMT https://communities.sas.com/t5/Statistical-Procedures/Regression-analysis-suggestion/m-p/71267#M3448 rockerd 2010-09-11T08:48:15Z https://communities.sas.com/t5/Statistical-Procedures/Regression-analysis-suggestion/m-p/71268#M3449 The only time statements like "var x1 has 20% var x2 has 12%" have meaning is when the variables are not correlated with one another.<BR /> <BR /> As soon as you have correlations, statements of the sort "var x1 has 20% var x2 has 12%" are meaningless.<BR /> <BR /> I suggest you adopt a different mindset. You can still determine which <I>combinations </I>of variables are important in the prediction equation; you cannot determine which individual variable is causing the impact via statistics alone, nor can you quantify the percent impact of an individual variable.<BR /> <BR /> One technique that allows you to determine the combinations of variables that are important in the predicting a response(s) is Partial Least Squares regression. Mon, 13 Sep 2010 13:08:24 GMT https://communities.sas.com/t5/Statistical-Procedures/Regression-analysis-suggestion/m-p/71268#M3449 Paige 2010-09-13T13:08:24Z