PROC REG: How do I create "outest" table with Partial Squares instead of Estimate parameters ?

[ekruger]

This should be simple. All I want to do is to create at table like OUTEST but instead of estimates, I want the partial rsquare of each explanatory variable in the table. Is there an easy way to do this?

[SteveDenham]

ODS OUTPUT is your friend.

Add the scorr1 and scorr2 options to your MODEL statement, and then use

ODS OUTPUT parameterestimates=parameterestimates;

and it will have the estimates, s.e.'s and partial squared correlations using Type I (sequential) and Type II (nonsequential) sums of squares.

Steve Denham

[ekruger]

Thanks, that worked perfectly!