In article <[EMAIL PROTECTED]>,
David Heiser <[EMAIL PROTECTED]> wrote:

>"Safa Gurcan" <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]
>> Dear Folks,
>> In multiple regression analysis; model summary (R2, adjusted R2) and ANOVA
>> results display in the output screen. I used ridge regression analysis to
>> my data in SYSTAT but, didn't see any of the results above except
>> standardized and unstandardized coefficient and lamda . Does Ridge
>> regression have R2 and SEM ? How de we understand that the unstandardized
>> regression coefficients explained the variation of dependent variable?

>> thanks for advice.

>> Safa Gurcan

>-------------------------------------
>One of the problems with ridge analysis is that there is no defined stopping
>point.  Ridge analysis is good when one wants to make "good" predictions of
>Y values where the X values  are near to and exceed the limits of the sample
>space. In many cases involving physical (chemical) data, the predictions
>exceed physical constraints, and by using ridge analysis, one "relaxes" the
>fit to where the predictions are withing "reality". Standard fit measures
>are not much help here to establish a "stopping point".

My paper in _Bayesian Statistics 3_ gives an explicit formula
for the loss in using a "wrong" ridge prior.  I suggest the
Bayesian approach be used rather than "classical" procedures
which are ad hoc with no real basis.
-- 
This address is for information only.  I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
[EMAIL PROTECTED]         Phone: (765)494-6054   FAX: (765)494-0558
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to