How do I use MVREGRESS function from the Statistics Toolbox with multi-dimensional output predictions Y?

I would like to perform multivariate linear regression in MATLAB. I have p predictors (independent variables) and n observations. My dependent variable Y is multidimensional, that is for every observation, my Y is a vector, not a scalar . In other words, my Y is an [n x d] matrix where d>1. I would like to know what options I have for performing this type of regression.

Answer 1

[MathWorks Support Team]

This is an example of how to use the MVREGRESS function in the case where your dependent variable Y is multidimensional. If your Y is single-dimensional, meaning that it is [n x 1], see the Related Solution at the bottom of this page.

Example using MVREGRESS with cell array inputs:

1. Download 'dataset.mat' which contains predictor and 3-dimensional target data.

2. Download and execute 'example2.m' which illustrates how to format the data into cell inputs to MVREGRESS. Comments in EXAMPLE2 provide additional information.

You may also wish to consult the documentation for the function MVREGRESS by typing 'doc mvregress' at the MATLAB command prompt.