How to add a categorical variable with a power of 2 to the rowexch function (D-opt design)

1 view (last 30 days)
Floris Bremmers
Floris Bremmers on 1 Sep 2014
Hello,
Given: A regression model with polynomial function 'poly42111'. The second factor has a power of 2 and is/can modeled as a continuous variable. However, the complete data set only has 4 levels of this factor.
Problem: To reduce the amount of data points required the model is subjected to a D-optimal design, by means of the rowexch function. To ensure the D-optimal design matrix contains available data only -- thus a selection of the four levels for the second factor -- the function is defined as follows:
dRE1 = rowexch(nfactors,points,'poly42111','categorical',2,'levels',4);
but now the following error occurs:
MODEL cannot specify powers for a categorical variable.
The rowexch function does not support categorical variables with power > 1. In fact, a simple regression model neither supports a factor with power >1 to be defined as categorical...
Question: Does anyone have a solution to this? different method / post-augmentation of this factor..
Help is much appreciated, thanks in advance!
Floris

Sign in to answer this question.

Answers (0)

Categories

Find more on Mathematics in Help Center and File Exchange

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!