search

pbreheny / grpreg

Regularization paths for regression models with grouped covariates
http://pbreheny.github.io/grpreg/
34 stars 14 forks source link

Weights option? #4

Open jeffwong-nflx opened 8 years ago

jeffwong-nflx commented 8 years ago

Is there any way to modify the code to accept weights? This would really help in making the package compatible with glmnet

pbreheny commented 8 years ago

I certainly recognize the value of this, but have never personally needed weights, so it remains low on my list of priorities. I'm going to tag this as an enhancement, because I do think it would be a useful feature, although I can't say I have any specific timeline for getting it done.

SlavichEve commented 1 year ago

I just wanted to follow this enhancement up. Increasingly regularly, I am coming across model selection with missing data problems where one approach is to fit a lasso on a stacked dataset of B imputed datasets, with weights equal to 1/B (or more complicated weights) to adjust for the inflated sample size. Currently unable to use grpreg which is my favourite group lasso package as without the weights too many variables are selected.