Download All

No BSD License

[beta shrcf rss]=nng_garotte(... Implementation of L. Breiman's non-negative garrote.
[beta,s,perror,se]=nng_finds(... Estimate 's' using either k-fold CV or the little bootstrap for the non-negative garotte.
err=nng_kcv(X,y,nfolds,s) k-fold cross validation
err=nng_little_bootstrap(X,y,... Implementation of the little bootstrap (Breiman 1995)
nng_example.m
View all files

Be the first to rate this file! 5 Downloads (last 30 days) File Size: 15.7 KB File ID: #20812

Non-negative Garotte

22 Jul 2008 (Updated 22 Jul 2008)

Implementation of Leo Breiman's non-negative garotte for linear regression

File Information
Description	Set of Matlab scripts for efficiently implementing Leo Breiman's non-negative garotte. For estimation of the garotte constraint we use either k-fold cross validation or the little bootstrap.
MATLAB release	MATLAB 7.4 (R2007a)

Tags for This File
Everyone's Tags	leo breiman, linear regression, nonnegative garotte, probability, regression, stai, statistics
Tags I've Applied
Add New Tags	Please login to tag files.

Comments and Ratings (2)
03 Aug 2008	Scott Miller	A very accessible summary of both cross-validation and the little bootstrap techniques used in this submission may be found on the web at: http://projecteuclid.org/DPubS?verb=Display&version=1.0&service=UI&handle=euclid.lnms/1215540960&page=record This is a very good review of model selection by C.R. Rao (one of the great statistical luminaries of the 20th century) and Y. Wu.
16 Aug 2008	Scott Miller	The neural network toolbox is needed to at least run the example, if not the program itself. Scott

Please login to add a comment or rating.