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.

.binary logistic regressioncategorical dataglmfitmultinomial logistic regressionnominal variables

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.

Best Answer

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

Related Solutions

MATLAB: Regression using 1×6 table with 500×20 double elements

I'm not sure I fully understand your question, but maybe this will help: If you want one alpha and five betas for each of the 500 rows separately, then you will have to call fitlm 500 times. Each time you call it, you should pass only the data for one row.

MATLAB: Surf() to bar3() “Matrix dimensions must agree, not rendering mesh”

That has to be the coolest simulation I’ve seen posted on MATLAB Answers!

I figured out how to make bar3 work.

1. Comment-out (or delete) this line ( 5 lines above the while loop ) and the axis line just below it:

%           h=surf(newH); % cannot change this to bar3

2. Comment out (or delete) this line in the while loop:

%               set(h,'Zdata', newH(i,j));

3. Insert these lines just below the commented-out ‘set(h, 'ZData' ...’ line:

                h = bar3(newH);
  %               h=surf(newH);
                axis([1 n 1 n -2 8]);

Then choose either bar3 or surf to see the result.

Best Answer

Related Solutions

MATLAB: Regression using 1×6 table with 500×20 double elements

MATLAB: Surf() to bar3() “Matrix dimensions must agree, not rendering mesh”

Related Question