Logistic Analysis
Examine Relationships between a Categorical Y and a Continuous X Variable
Use the Logistic platform to fit a logistic regression model to a categorical Y variable with a continuous X variable. You can view ROC curves, lift curves, and odds ratio estimates. The fitted model provides estimated probabilities for each value of the X variable. You can also perform inverse prediction, which enables you to predict the X value for a specific probability value of the Y variable.
The Logistic platform is the nominal or ordinal by continuous personality of the Fit Y by X platform. There is a distinction between nominal and ordinal responses in this platform:
• Nominal logistic regression models estimate a set of curves that partition the probability among the levels of a nominal response variable. An example of a nominal logistic regression model is shown on the right side of Figure 8.1.
• Ordinal logistic regression models estimate the probability of being less than or equal to a target level of an ordinal response variable. The model estimates a single logistic curve that gets shifted horizontally to produce probabilities for the ordered categories. This model is less complex and is recommended for ordered responses. An example of an ordinal logistic regression model is shown on the left side of Figure 8.1.
Figure 8.1 Examples of Ordinal and Nominal Logistic Regression
