Overview of the Logistic Regression Model

A logistic regression attempts to predict the value of a binary response variable. A logistic regression analysis models the natural logarithm of the odds ratio as a linear combination of the explanatory variables. This approach enables the logistic regression model to approximate the probability that an individual observation belongs to the level of interest.

The logistic regression model requires a category response variable and at least one effect variable or interaction term. When your category response variable contains more than two levels, SAS Visual Statistics prompts you to select the level of interest. That is, SAS Visual Statistics treats all observations in the level of interest as an event and all other observations as nonevents.