MGLM

MGLM Toolbox for Matlab

MGLM toolbox is a collection of Matlab functions for multi-response GLM regression and sparse regression.

The toolbox is developed by Hua Zhou and Yiwen Zhang. An R package of same functionalities is available at CRAN.

Compatibility

The code is tested on Matlab R2017a, but should work on other versions of Matlab with no or little changes. Current version works on these platforms: Windows 64-bit, Linux 64-bit, and Mac (Intel 64-bit). Type computer in Matlab’s command window to determine the platform.

Installation (Matlab version >= 2014b)

Download the Matlab toolbox installation file MGLM.mltbx. Double click the downloaded file and you should be good to go. If it does not work for some reasons, follow the below instructions for Matlab version < 2014b.

Installation (Matlab version < 2014b)

1. Download ZIP File file using the links on the left.
2. Extract the zip file.

   unzip Hua-Zhou-MGLM-xxxxxxx.zip
   

3. Rename the folder from Hua-Zhou-MGLM-xxxxxxx to MGLM.

   mv Hua-Zhou-MGLM-xxxxxxx MGLM
   

4. Add the MGLM folder to Matlab search path. Start Matlab, cd to the MGLM directory, and execute the following commands
   addpath(pwd) %<-- Add the toolbox to the Matlab path
save path %<-- Save for future Matlab sessions
5. Go through following tutorials for the usage. For help of individual functions, type ? followed by the function name in Matlab.

Tutorial

• Dirichlet-Multinomial distribution
• Generalized Dirichlet-Multinomial distribution
• Negative multinomial distribution
• Multinomial-logit regression and sparse regression
• Dirichlet-Multinomial regression and sparse regression
• Generalized Dirichlet-Multinomial regression and sparse regression
• Negative multinomial regression and sparse regression

How to cite

If you use this toolbox in any way, please cite the software itself along with at least one publication or preprint.

• Software reference
  H Zhou and Y Zhang. Matlab MGLM Toolbox Version 1.0.0, Available online, March 2017.
• Default article to cite for the toolbox
  Y Zhang, H Zhou, J Zhou, and W Sun. (2017) Regression models for multivariate count data. Journal of Computational and Graphical Statistics, 26(1):1-13. [pdf]

Contacts

Hua Zhou huazhou@ucla.edu.