Comments on: Tues Mar 27 http://www.stat.cmu.edu/~kass/smnp/?p=109 Thu, 26 Apr 2012 14:07:02 +0000 hourly 1 http://wordpress.org/?v=3.0.4 By: Kelly http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-242 Kelly Tue, 27 Mar 2012 13:30:55 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-242 I'm still unclear on how cross-validation works. Can you clarify? I’m still unclear on how cross-validation works. Can you clarify?

]]> By: Thomas Kraynak http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-241 Thomas Kraynak Tue, 27 Mar 2012 13:30:52 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-241 If you want to simplify your model, what are some methods to determine whether to use forward selection, backward elimination, and/or stepwise regression? If you want to simplify your model, what are some methods to determine whether to use forward selection, backward elimination, and/or stepwise regression?

]]> By: Matt Bauman http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-240 Matt Bauman Tue, 27 Mar 2012 13:27:13 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-240 I'm slightly confused by the remarks on shrinkage. Why does a high coefficient suggest an irrelevant variable? How does adding the penalty term fix problems with non-invertible matrices? I’m slightly confused by the remarks on shrinkage. Why does a high coefficient suggest an irrelevant variable? How does adding the penalty term fix problems with non-invertible matrices?

]]> By: Rex Tien http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-239 Rex Tien Tue, 27 Mar 2012 13:15:44 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-239 It would be very helpful to have a picture of the example in 12.5.5. Is there any way of selectively shrinking certain coefficients selectively? It would be very helpful to have a picture of the example in 12.5.5.

Is there any way of selectively shrinking certain coefficients selectively?

]]> By: Rob Rasmussen http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-238 Rob Rasmussen Tue, 27 Mar 2012 13:05:32 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-238 Is it possible to use a form of multiple linear regression for non-cosine directional tuning, or would it be limited to using kernel regression methods or some other technique? Is it possible to use a form of multiple linear regression for non-cosine directional tuning, or would it be limited to using kernel regression methods or some other technique?

]]> By: Yijuan Du http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-237 Yijuan Du Tue, 27 Mar 2012 12:33:37 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-237 When we chosee the explanatory variables, do we usually try to make them uncorrelated or it does not matter whether they are correlated? The correlation of them only influences interpretation? When we chosee the explanatory variables, do we usually try to make them uncorrelated or it does not matter whether they are correlated? The correlation of them only influences interpretation?

]]> By: David Zhou http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-236 David Zhou Tue, 27 Mar 2012 07:40:54 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-236 Can you go over the source localization problem for MEG that you go into in Example 12.9? I'm not clear on your use of penalized least squares in minimum norm estimate. It sounds like an exciting statistical technique, but I'd love to hear more about it. Can you go over the source localization problem for MEG that you go into in Example 12.9? I’m not clear on your use of penalized least squares in minimum norm estimate. It sounds like an exciting statistical technique, but I’d love to hear more about it.

]]> By: Matt Panico http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-235 Matt Panico Tue, 27 Mar 2012 06:30:53 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-235 Does the forward selection algorithm also apply to interaction effects? Should we consider interaction effects with every variable we add to the model, or only where we think it would make sense? Does the forward selection algorithm also apply to interaction effects? Should we consider interaction effects with every variable we add to the model, or only where we think it would make sense?

]]> By: Jay Scott http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-234 Jay Scott Tue, 27 Mar 2012 05:35:47 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-234 Regarding model selection, will 'forward selection' and 'backward elimination' converge on the same model? Regarding model selection, will ‘forward selection’ and ‘backward elimination’ converge on the same model?

]]> By: Ben Dichter http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-233 Ben Dichter Tue, 27 Mar 2012 04:29:08 +0000 http://www.stat.cmu.edu/~kass/smnp/?p=109#comment-233 When looking at interaction effects, you might end up with a very large number of variables. How do you know which combinations of variables might be important? How do you deal with over-fitting? When looking at interaction effects, you might end up with a very large number of variables. How do you know which combinations of variables might be important? How do you deal with over-fitting?

]]>