Message-ID: <6731304c0903170600p25dc5eeaneb550df2eafa7e9f@mail.gmail.com>
Date: 2009-03-17T13:00:52Z
From: Max Kuhn
Subject: Double Cross validation for LASSO
In-Reply-To: <8bdeb1b50903170531v2e096027w8558904e57404b13@mail.gmail.com>
Alex,
> ?I am looking for a code on double cross validation in
> LASSO , one for optimizing the parameter and other one is for MSEP. If any
> one have it, please foroward to me. I am using different ?package like LARS,
> chemometric etc.
It won't do double cross-validation, but the caret package can be used
to tune similar models using different forms of cross-validation or
the bootstrap. Currently, the elasticnet, the lasso (via the
elasticnet) and models from the penalized package can be used. I have
a devel version with glmnet in it too. I could always add LARS to
caret. See
http://cran.r-project.org/web/packages/caret/index.html
http://www.jstatsoft.org/v28/i05
Of course, there is always lars:::cv.lars too.
--
Max