s-news
[Top] [All Lists]

Mandatory variable inclusion with LASSO?

To: "'S-PLUS Newsgroup'" <s-news@lists.biostat.wustl.edu>
Subject: Mandatory variable inclusion with LASSO?
From: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 26 Feb 2009 17:10:21 -0500
In-reply-to: <0A86EF38B23C4ECB81BD47ECB0A98D0C@prosanos.local>
References: <40E02871B8B84F4FB46FA9F3D898B9DA010FD82E@mb-exch.ensco.win> <0A86EF38B23C4ECB81BD47ECB0A98D0C@prosanos.local>
Thread-index: AcmTpqHOsHnhvenwQLKIPGA1nR0bMwDpCjpgAERSGvA=

I am familiarizing myself with LASSO for Poisson regression using the glars package to model discrete outcomes in observational healthcare data sets.  The technique performs as advertized in terms of selecting reasonable subsets of variables for inclusion in the models, picking from the large number of variables that are available to me.  I have, however, run into two issues:

 

1)       I would like to set things up so that a particular variable, namely treatment with Drug X, is always included in the model.  I can do this with stepwise regression [ step() ] by specifying treatment in my “lower” model, but I don’t know how to do it with the available LASSO package.  Can this be done?

2)       I am more interested in getting an accurate coefficient estimate for the treatment variable than I am in overall prediction accuracy.  This coefficient is “my answer”.  The LASSO as currently configured will shrink the treatment coefficient so that there is some headroom within the constraint box to share with other variables that improve overall accuracy.  Is there a way to remove the penalty constraint from one particular variable, while keeping it for the others?  Does it make sense to do this?

 

Thanks in advance,

 

Alan

 

Alan Hochberg

VP, Research

ProSanos Corporation

225 Market St. Ste. 502,

Harrisburg, PA 17101

Tel 717-635-2124 * Fax 717-635-2575

 

<Prev in Thread] Current Thread [Next in Thread>