Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

New to MATLAB?

I have one binary response variable and 5 nominal predictor variables which take values (strongly agree, agree, neutral, disagree, strongly disagree). I want to obtain coefficient matrix using logistic regression. Please help.

Asked by jacksparrow

jacksparrow (view profile)

on 8 Apr 2013

Y = [0 1 0 1 1 1 ..] X = [var1 var2 var3 var4 var5] var1 = [agree neutral disagree ...] similarly var 2 var 3 var4 var5 I converted responses to -2 -1 0 1 2 as strongle disagree, disagree, neutral etc. now,I ran: [b,dev,stats] = glmfit(X,Y,'binomial','link','logit') I am getting error as: Y must be a two column matrix or a vector for the binomial distribution.

0 Comments

jacksparrow

jacksparrow (view profile)

Products

No products are associated with this question.

1 Answer

Answer by Tom Lane

Tom Lane (view profile)

on 9 Apr 2013
Accepted answer

Make sure Y and var1,...var5 are column vectors.

3 Comments

jacksparrow

jacksparrow (view profile)

on 11 Apr 2013

Thanks! mnrfit worked for me. As it turns out, I was using n x 1 column matrix instead of n x 2.

jacksparrow

jacksparrow (view profile)

on 11 Apr 2013

Would you be kind enough to provide me with a link to understand the stats matrix parameters to help me validate the results of my analysis.

Tom Lane

Tom Lane (view profile)

on 14 Apr 2013

You can start with this:

http://www.mathworks.com/help/stats/mnrfit.html

Then if you have a specific question, do let me know.

Tom Lane

Tom Lane (view profile)

Contact us