variable selection among highly correlated variables

Go down

variable selection among highly correlated variables

Post by Debojyoti Chakraborty on Fri Nov 25, 2016 11:11 am

Dear colleagues
You may remember our discussion during the SUSTREE workshop regarding the tricky issue of variable selection. As I mentioned I would like to share this paper(Gregorutti et al. 2016) which elaborates the efficient Recursive feature Elimination algorithm with Random forest.

This process is very efficient and was found to perform better than the traditional stepwise selection in multivariate statistics and the commonly used variable importance(reducing MSE and GINI index)within the random Forest

best regards

Debojyoti Chakraborty

Posts : 2

Back to top Go down

Back to top

- Similar topics

Permissions in this forum:
You cannot reply to topics in this forum