This version of the Help is no longer updated. See JMP.com/help for the latest version.

.
Fitting Linear Models > Logistic Regression Models > Overview of the Nominal and Ordinal Logistic Personalities
Publication date: 07/30/2020

Overview of the Nominal and Ordinal Logistic Personalities

Logistic regression models the probabilities of the levels of a categorical Y response variable as a function of one or more X effects. The Fit Model platform provides two personalities for fitting logistic regression models. The personality that you use depends on the modeling type (Nominal or Ordinal) of your response column.

For more information about fitting logistic regression models, see Walker and Duncan (1967), Nelson (1976), Harrell (1986), and McCullagh and Nelder (1989).

For more information about the parameterization of the logistic regression model, see Logistic Regression Model.

Nominal Logistic Regression

When the response variable has a nominal modeling type, the platform fits a linear model to a multi-level logistic response function using maximum likelihood. Therefore, all but one response level is modeled by a logistic curve that represents the probability of the response level given the value of the X effects. The probability of the final response level is 1 minus the sum of the other fitted probabilities. As a result, at all values of the X effects, the fitted probabilities for the response levels sum to 1.

If the response variable is binary, you can set the Target Level in the Fit Model window to specify the level whose probability you want to model. By default, the model estimates the probability of the first level of the response variable.

For more information about fitting models for nominal response variables, see Nominal Responses in the Statistical Details section.

Ordinal Logistic Regression

When the response variable has an ordinal modeling type, the platform fits the cumulative response probabilities to the logistic function of a linear model using maximum likelihood. Therefore, the cumulative probability of being at or below each response level is modeled by a curve. The curves are the same for each level except that they are shifted to the right or left.

Tip: If there are many response levels, the ordinal model is much faster to fit and uses less memory than the nominal model.

For more information about fitting models with ordinal response variables, see Ordinal Responses in the Statistical Details section.

Other JMP Platforms That Fit Logistic Regression Models

There are many other places in JMP where you can fit logistic regression models:

To fit logistic regression models with a single continuous main effect, you can use the Fit Y by X platform to see a cumulative logistic probability plot for each effect. See Logistic Analysis in Basic Analysis.

To perform variable selection in logistic regression models, you can use the Stepwise personality of the Fit Model platform. See Stepwise Regression Models.

To fit logistic regression models that use a link function other than the Logit link, you can use the Generalized Linear Model personality of the Fit Model platform. See Generalized Linear Models.

To perform variable selection in logistic regression models and fit penalized logistic regression models, you can use the Generalized Regression personality of the Fit Model platform. See Generalized Regression Models.

Want more information? Have questions? Get answers in the JMP User Community (community.jmp.com).