Comments on: Linear regression http://www.spsslog.com/2006/08/21/linear-regression/ From users, for users Wed, 07 Jan 2009 13:53:04 +0000 http://wordpress.org/?v=2.7 hourly 1 By: andris http://www.spsslog.com/2006/08/21/linear-regression/comment-page-1/#comment-159 andris Wed, 13 Dec 2006 08:48:55 +0000 http://www.spsslog.com/2006/08/21/linear-regression/#comment-159 Yves, Dave, Thank you for you additions! This make the answer more complete. Yves, Dave,

Thank you for you additions! This make the answer more complete.

]]> By: Dave in Canada http://www.spsslog.com/2006/08/21/linear-regression/comment-page-1/#comment-157 Dave in Canada Tue, 12 Dec 2006 21:52:02 +0000 http://www.spsslog.com/2006/08/21/linear-regression/#comment-157 Andris' explanation of linear regression is correct but very brief. Yves added the important step of checking the outliers. I want to remind users that Linear regression (the regression which Andris explains) only works if the dependent variable -- "costs" in Andris' example -- is a continuous variable. The variable does not have to be perfectly continuous. Some say that you can have as few as seven possible levels of outcome in the dependent variable. But with fewer than that, the linear regression method begins to give misleading answers. <strong>The alternative is in an add-on module from SPSS, called Regression Models. It has methods called "Binary Logistic" and "Muiltinomial Logistic" regression. Use these when the dependent variable is binary, like Yes/No or Passed/Failed; or when it is multinomial, like Low-Medium-High.</strong> Andris’ explanation of linear regression is correct but very brief. Yves added the important step of checking the outliers. I want to remind users that Linear regression (the regression which Andris explains) only works if the dependent variable — “costs” in Andris’ example — is a continuous variable. The variable does not have to be perfectly continuous. Some say that you can have as few as seven possible levels of outcome in the dependent variable. But with fewer than that, the linear regression method begins to give misleading answers.

The alternative is in an add-on module from SPSS, called Regression Models. It has methods called “Binary Logistic” and “Muiltinomial Logistic” regression. Use these when the dependent variable is binary, like Yes/No or Passed/Failed; or when it is multinomial, like Low-Medium-High.

]]> By: Yves http://www.spsslog.com/2006/08/21/linear-regression/comment-page-1/#comment-127 Yves Tue, 21 Nov 2006 19:44:45 +0000 http://www.spsslog.com/2006/08/21/linear-regression/#comment-127 Regression analysis is not conducted right this way. You have to make sure you have a look at the outliers, at the normal distribution, at the scatterplot zpred*zresid, and so on... Only when you take notice of these "problems", your regression will be valuable Regression analysis is not conducted right this way. You have to make sure you have a look at the outliers, at the normal distribution, at the scatterplot zpred*zresid, and so on…

Only when you take notice of these “problems”, your regression will be valuable

]]> By: andris http://www.spsslog.com/2006/08/21/linear-regression/comment-page-1/#comment-82 andris Mon, 18 Sep 2006 10:17:12 +0000 http://www.spsslog.com/2006/08/21/linear-regression/#comment-82 Franklin, see <a href="http://en.wikipedia.org/wiki/Residual_sum_of_squares">Wikipedia</a> Franklin, see Wikipedia

]]> By: Franklin http://www.spsslog.com/2006/08/21/linear-regression/comment-page-1/#comment-81 Franklin Mon, 18 Sep 2006 07:45:27 +0000 http://www.spsslog.com/2006/08/21/linear-regression/#comment-81 What dothe residual sum of suqares mean What dothe residual sum of suqares mean

]]>