RapidMiner

Regularisation for Logistic Regression/ Feature Selection

Contributor

Regularisation for Logistic Regression/ Feature Selection

Hello, I was trying to do regularisation for logistic regression and when the output came out, the p-values were all question marks in the cells.

 

When I don't do regularisation, I am able to get the p-values.

 

Please help?

2 REPLIES
Community Manager

Re: Regularisation for Logistic Regression/ Feature Selection

Yep, I see this too. This might be a bug. Let me track down development.

Regards,
Thomas - Community Manager
LinkedIn: Thomas Ott
Highlighted
Moderator

Re: Regularisation for Logistic Regression/ Feature Selection

Hi,

it's no bug. it's not part of the h2o package. see: http://docs.h2o.ai/h2o/latest-stable/h2o-docs/data-science/algo-params/compute_p_values.html

 

I think it also makes no sense to calculate p-values if you add regularization (=bias).

 

~Martin

--------------------------------------------------------------------------
Head of Data Science Services at RapidMiner