s-news
[Top] [All Lists]

Re:

To: "Fan Xiao" <xiaofan@pilot.msu.edu>, <s-news@wubios.wustl.edu>
Subject: Re:
From: Marc Feldesman <feldesmanm@pdx.edu>
Date: Wed, 28 Feb 2001 07:53:12 -0800
In-reply-to: <000501c0a145$10025ea0$90b0b718@c1337008A>
At 12:12 AM 2/28/2001 -0500, Fan Xiao wrote:
hello
would you please tell me if there is a function that can do cross-validation?
thanks.

What are you trying to cross-validate? The bootstrapping functions in S-Plus, the various bootstrapping add-on libraries (e.g. boot), and many of the multivariate routines (discrim, lda, qda, rpart) all have capabilities to cross-validate. Do you want k-fold, leave-one-out?

There is no generic function called "cross-validate". I'd suggest that you read the S-Plus manuals on bootstrapping and the Venables and Ripley book "Modern Applied Statistics with S-Plus, 3rd Edition, 1999) for further discussion.


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