RE: st: RE: RE: RE: LARS ado??

["Ploutz-Snyder, Robert (JSC-SK)[USRA]" <robert.ploutz-snyder-1@nasa.gov>]

Peter--
Great info--thanks much for forwarding.

For now I guess R is the solution, but my hope is that Stata will consider writing LASSO or LARS routines, since they have already gained momentum among the biostatistics/biometrics communities, and the horizon points to more advances here.

Rob

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Lachenbruch, Peter
Sent: Thursday, April 01, 2010 10:10 AM
To: 'statalist@hsphsun2.harvard.edu'
Subject: RE: st: RE: RE: RE: LARS ado??

I wrote Trevor Hastie and he responded as follows:

Hi Tony

There is the glmpath package in R that is available, as well
as the newer glmnet package. 
The latter is very fast, and the next version due for release soon has Poisson and Cox included.

best
Trevor

So the answer is that if you want LASSO or LARS for logistic you need to visit R.  
 
Tony

Peter A. Lachenbruch
Department of Public Health
Oregon State University
Corvallis, OR 97330
Phone: 541-737-3832
FAX: 541-737-4001

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ploutz-Snyder, Robert (JSC-SK)[USRA]
Sent: Wednesday, March 31, 2010 2:18 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: RE: RE: RE: LARS ado??

Thank you much Peter. I appreciate your followup.

On Mar 31, 2010, at 2:58 PM, "Lachenbruch, Peter" <Peter.Lachenbruch@oregonstate.edu 
 > wrote:

> In reading the original LARS article, the derivation was only for  
> OLS.  Sorry I missed your logit interest.  I asked Brad Efron about  
> the extension and he mentioned that Trevor Hastie and Rob Tibshirani  
> may be working on something, but I haven't heard anything on it.
>
> Tony
>
> Peter A. Lachenbruch
> Department of Public Health
> Oregon State University
> Corvallis, OR 97330
> Phone: 541-737-3832
> FAX: 541-737-4001
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner- 
> statalist@hsphsun2.harvard.edu] On Behalf Of Ploutz-Snyder, Robert  
> (JSC-SK)[USRA]
> Sent: Wednesday, March 31, 2010 8:28 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: RE: LARS ado??
>
> Thanks Tony,
> That email bounced back to me this week.
> And the LARS routine as written is for typical ols regression  
> context, I believe, not Logit (binary outcome).
> Rob
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner- 
> statalist@hsphsun2.harvard.edu] On Behalf Of Lachenbruch, Peter
> Sent: Wednesday, March 31, 2010 10:19 AM
> To: 'statalist@hsphsun2.harvard.edu'
> Subject: st: RE: LARS ado??
>
> The LARS routine has a LASSO  option.  The email at the bottom of  
> the help file was
> adrian.mander@mrc-hnr.cam.ac.uk
> It worked when I tried it a year or so ago.
>
>
> Tony
>
> Peter A. Lachenbruch
> Department of Public Health
> Oregon State University
> Corvallis, OR 97330
> Phone: 541-737-3832
> FAX: 541-737-4001
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner- 
> statalist@hsphsun2.harvard.edu] On Behalf Of Ploutz-Snyder, Robert  
> (JSC-SK)[USRA]
> Sent: Wednesday, March 31, 2010 7:04 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: LARS ado??
>
> Is anyone aware of a Stata ado for running LASSO variable selection  
> techniques to compliment the LARS.ado that Adrian Mander has begun?   
> Specifically a routine written for logit models?
>
> Also--the email addy that Adrian provides in his help file bounces  
> back.  Anyone know how to reach Adrian Mander?
>
> Thanks much,
> Rob
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/