Home >> Science & Nature >> Science

How to Graph Predicted Probabilities in SAS

Logistic regression is a statistical technique for looking at the relationship between a dichotomous dependent variable and one or more independent variables. For example, you might look at the relationship between being a Democrat or Republican and variables such as age, race/ethnicity and income. Statistical Analysis Software, or SAS, lets you fit logistic regressions and graph the predicted probabilities from a model.

Instructions

- 1
  Start ODS graphics. Before SAS can display graphs, you must tell it that graphics are coming up by putting ODS graphics on in your SAS program.
- 2
  Write a PROC LOGISTIC model as you ordinarily would. In the example, this might look like
  
  PROC LOGISTIC DATA = MYDATA;
  
  CLASS RACE;
  
  MODEL PARTY = RACE INCOME AGE;
- 3
  Add an EFFECTPLOT statement and run the program. The example would become
  
  PROC LOGISTIC DATA = MYDATA;
  
  CLASS RACE;
  
  MODEL PARTY = RACE INCOME AGE;
  
  EFFECTPLOT;
  
  RUN;

Science