Re: Regression model under-predicts high values of Y? What can I do?

On 17 Jan 2007 18:36:57 -0800, pbrewster@xxxxxxxxxxx wrote:

HI all, I'm modeling a DV that ranges between 1 and 30, with a mean of
11.38 and standard deviation of 5.042. My model are consistently
under-predicting the higher values of Y (highest Y hat is 18), which
incidentally are the cases I am most interested in predicting correctly
(most important according to my theory. I have tried taking the
natural log of the DV, and this does not help. Does anyone have
suggestions on what I can do to improve my explanatory power on the
higher values of Y?

This is about what you would expect to happen. The highest Y's have
the largest errors, and you can't predict the errors.

That's assuming the relation is linear. You might well try adding x^2
or other transformations to pick up nonlinearity.
-*** Startz
.