Hi, I'm getting this error in ANOVAN: Undefined function or variable "denommat". I have 3 factors, one is a random effect (which individual), the others are predictors.
Y = randn(100,1); subject = repmat([1:10]', 10,1); traits = randn(100,2); [p,T,stats] = anovan(Y, [subject, traits], ... 'varnames', {'Subject','Trait1', 'Trait2'} , ... 'random', 1, ... 'continuous', [2 3], ... 'model', [0 1 0; 0 0 1; 0 1 1] ... );
- I'm looking for the effects of the predictors and their interaction.
- But it's a mixed effects design, in that, I have repeated measures of the traits for each subject. I've therefore specified 'Subject' as a random effect. What I intend here is to account for the fact that repeated trait measures on an individual may correlated.
- I think the problem is that the "subject" factor is not used in the design matrix: I've used a column of zeros to specify that the subject is not a categorical regressor (if it were, it would eat up all the between-subject variability, right?)
Best Answer