topic Multiple Logistic Regression Code Help in Statistical Procedures

Multiple Logistic Regression Code Help

steph_ — Thu, 21 Nov 2024 17:30:18 GMT

Hello!

I am trying to write code for multiple logistic regression and cannot find what code to use.

My outcome/y variable is binary (yes or no but recoded to be numeric) and explanatory/x variables are mainly continuous but there is one binary (recoded to be numeric) variable too.

Could someone assist me with some basic code for a starting point?

Thanks

Re: Multiple Logistic Regression Code Help

StatDave — Thu, 21 Nov 2024 17:37:15 GMT

Very simple for basic model fit. Assuming your response is Y with values 0 or 1 where 1 is the event of interest and your data set is named MYDATA:

proc logistic data=mydata;

class <list any categorical predictors> / param=glm;

model y(event='1') = <list all predictors, continuous and categorical>;

run;

See other examples in the PROC LOGISTIC documentation - Getting Started and Examples sections. There are many options for additional information.

Re: Multiple Logistic Regression Code Help

steph_ — Thu, 21 Nov 2024 17:51:56 GMT

Thank you! What does the param=glm mean?

Re: Multiple Logistic Regression Code Help

StatDave — Thu, 21 Nov 2024 17:55:49 GMT

Again, see that documentation for info on all options, but PARAM=GLM creates the typical 0,1 coded design columns that most people use to represent categorical predictors. But if your only categorical predictors are binary and you already have them coded as 0,1, then you can just omit the CLASS statement.