Is it possible to evaluate several observation with one mnrval call?

In the .m file I can read the following: PHAT = MNRVAL(B,X) computes predicted probabilities for the nominal multinomial logistic regression model with predictor values X. X is an N-by-P design matrix with N observations on P predictor variables.

Previously I used mnrval(B,x), where x was a single observation vector, but is it possible to evaluate x1 and x2 at the same time by creating an NxP observation matrix? This is not working for me! mnrval(B,x1) and mnrval(B,x2) is OK, but mnrval(B,[x1;x2]) is not working for me, nor mnrval(B,[x1;x2]'). Why?

Answer 1

[Soma Ardhanareeswaran]

The sizes of B and X may be incompatible. Please make sure that X contains n observations for p predictors.

It is not possible to evaluate several observations in a single call.