Automatically exported from code.google.com/p/envlp
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Data
data
doc
examples
src
.hgignore
Contents.m
README.txt
envlp_license.m
install_envlp.m

README.txt

MATLAB TOOLBOX: envlp

VERSION:
1.0

DATA:
04-Sep-2012

LICENSE:
GPL-2

AUTHORS:
Dennis Cook (dennis@stat.umn.edu)
Zhihua Su (zhihuasu@stat.ufl.edu)  
Yi Yang (yiyang@umn.edu)

URL:
http://code.google.com/p/envlp/


Description
___________

A MATLAB Toolbox for Computing Envelope Estimators in Multivariate
Linear Regression.

The envlp toolbox implements a variety of models in the envelope area
under the framework of multivariate linear regression or estimating the
multivariate mean:

1.  envelope model
2.  partial envelope model
3.  heteroscedastic envelope model
4.  inner envelope model
5.  scaled envelope model
6.  envelope model in the predictor space
7.  envelope model for estimating the multivariate mean

Installation
____________

You can place the envlp directory anywhere. To install the toolbox,
direct your MATLAB working directory to the folder `envlp/`, and type
`install_envlp` If the users agree with our license statements, the
installation is completed, and all the utilities of the toolbox are
added to the MATLAB path.

NOTE: If a previous version of the toolbox is present, it should be
removed before installing the new version. To remove, simply delete the
folder of the toolbox.

Documentation
_____________

You can find a very comprehensive user's guide and reference manual in
the `envlp/doc/` subdirectory.

Examples and Datasets
_____________________

Trying some examples in the `envlp/examples/` subdirectory is a good way
to check if envlp is installed properly. Running these examples requires
datasets in `envlp/data/` subdirectory. See `envlp/data/README_data.txt`
for further information.

Technical support
_________________

We provide a support website, on which the users can download the
toolbox, report bugs and check recent updates
http://code.google.com/p/envlp/. For further help, the users can contact
the authors of the toolbox: Dennis Cook (dennis@stat.umn.edu), Zhihua Su (zhihuasu@stat.ufl.edu) and Yi Yang (yiyang@umn.edu).

References
__________

envlp: A MATLAB Toolbox for Computing Envelope Estimators in
Multivariate Linear Regression.
Dennis Cook, Zhihua Su, Yi Yang.
Technical report (2012)