Timeline for Model stability when dealing with large $p$, small $n$ problem

9 events

when toggle format	what		by	license	comment
Jun 1, 2012 at 15:56	comment	added	Pardis		According to the authors it is also an improvement over elastic net.
Jun 1, 2012 at 15:54	history	edited	Pardis	CC BY-SA 3.0	added 92 characters in body
Jun 1, 2012 at 15:42	comment	added	Pardis		I have added another link that I think answers your question better.
Jun 1, 2012 at 15:42	history	edited	Pardis	CC BY-SA 3.0	added 1284 characters in body
Jun 1, 2012 at 9:37	comment	added	dimi		I imagine, I can similarly use randomized elastic net in this case.
Jun 1, 2012 at 9:36	comment	added	dimi		What happens under approximate collinearity between the explanatory variables? In the conventional forward search algorithm in regression analysis, we are often faced with the situation where two variables x1 and x2 have similar explanatory power. If x1 is in the model, then there is no need to include x2; conversely, if x2 is in the model there is no need to include x1. If I understand your procedure correctly, you will tend to include x1 half the time and x2 half the time, leading to stability probabilities of about 50% each. If so, you might falsely conclude that neither variable is needed.
Jun 1, 2012 at 9:28	comment	added	dimi		Thank you for remarks. I have also considered randomized lasso, but would it suit in the case of collinearity?
May 31, 2012 at 21:13	comment	added	jbowman		Nice links(+1).
May 31, 2012 at 20:13	history	answered	Pardis	CC BY-SA 3.0