ridge regression with correlation/covariance input

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

ridge regression with correlation/covariance input

Maguin, Eugene

Spss provides a ridge regression macro that uses the raw data file. We’ve passed the raw data through mplus to get a missing data included covariance matrix and we want to pass that through a ridge regression procedure. Has anyone done ridge regression with matrix input? Practically, I think this means having modified the ridge regression macro to process a covariance matrix?

 

Thanks, Gene Maguin

Reply | Threaded
Open this post in threaded view
|

Re: ridge regression with correlation/covariance input

David Marso
Administrator
If you look at the innards, one of the first thing it does is create a correlation matrix.
Just modify the macro to read the correlation matrix instead of the raw data.
You can convert the COVARIANCE matrix to a correlation matrix using MCONVERT.

Maguin, Eugene wrote
Spss provides a ridge regression macro that uses the raw data file. We've passed the raw data through mplus to get a missing data included covariance matrix and we want to pass that through a ridge regression procedure. Has anyone done ridge regression with matrix input? Practically, I think this means having modified the ridge regression macro to process a covariance matrix?

Thanks, Gene Maguin
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"