multinomial logistic regression

stellapersis7 — Sat, 11 Nov 2023 05:27:05 GMT

Hi all,

I am trying to do a multinomial logistic regression for a study with 3 categories dependent variable (SDMSCORE) and 4 categories independent variable (REGIONEW). I have used the code below. (its MEPS data .. survey based, so used surveylogistic)

proc surveylogistic data = meps.sdm;

weight poolwt;

cluster VARPSU;

strata VARSTR;

class agenew;

model SDMSCORE = regionnew / link =glogit;

run;

So in my results, it has taken a reference value in both dependent (poor) and independent(category 4) variable. IS that correct. I was thinking one of the categories in independent variable is taken as reference. the results are below:

Re: multinomial logistic regression

SAS_Rob — Mon, 13 Nov 2023 15:55:02 GMT

Yes, you will get n-1 sets of coefficients, where n is the number of levels to the response variable. This is because they are restricted in a sense that the probabilities must sum to 1. There is a good explanation as to how to interpret a generalized logit model in general in the LOGISTIC documentation.

SAS Help Center

topic multinomial logistic regression in Statistical Procedures

multinomial logistic regression

Re: multinomial logistic regression