Publication date: 07/30/2020

Fitting Linear Models

Using the Fit Model platform, you can specify complex models efficiently. Your task is simplified by Macros, Attributes, and transformations. Fit Model is your gateway to fitting a broad variety of models and effect structures.

These include:

• simple and multiple linear regression

• analysis of variance and covariance

• random effect, nested effect, mixed effect, repeated measures, and split plot models

• nominal and ordinal logistic regression

• multivariate analysis of variance (MANOVA)

• canonical correlation and discriminant analysis

• loglinear variance (to model the mean and the variance)

• generalized linear models (GLM)

• parametric survival and proportional hazards

• response screening, for studying a large number of responses

In JMP Pro, you can also fit the following:

• mixed models with a range of covariance structures

• generalized regression models including the elastic net, lasso, and ridge regression

• partial least squares

The Fit Model platform lets you fit a large variety of types of models by selecting the desired personality. This chapter focuses on the elements of the Model Specification window that are common to most personalities.

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