The Fit Model platform provides an efficient way to specify models that have complex effect structures. These effect structures are linear in the model parameters. Once you have specified your model, you can select the appropriate fitting technique from a number of fitting personalities. Once you choose a personality, the Fit Model window provides choices that are relevant for the chosen personality. This chapter focuses on the elements of the Model Specification window that are common to most personalities. For a description of all personalities, see Elements in the Fit Model Launch Window.
Fit Model can be used to specify a wide variety of models that can be fit using various methods. Table 2.1 lists some typical models that can be defined using Fit Model. In the table, the effects X and Z represent columns with a continuous modeling type, while A, B, and C represent columns with a nominal or ordinal modeling type.
See Model Specification Templates for the clicking sequences that produce these model effects, plots of the model fits, and some examples.
|
Type of Model |
Model Effects |
|---|---|
|
Simple Linear Regression |
X |
|
Polynomial in X to Degree k |
X, X*X,..., Xk |
|
Polynomial in X and Z to Degree k |
X, X*X,..., Xk, Z, Z*Z,..., Zk |
|
Multiple Linear Regression |
X, Z, and other continuous columns |
|
One-Way Analysis of Variance |
A |
|
Two-Way Analysis of Variance |
A, B |
|
Two-Way Analysis of Variance with Interaction |
A, B, A*B |
|
Three-Way Full Factorial |
A, B, C, A*B, A*C, B*C, A*B*C |
|
Analysis of Covariance, Equal Slopes |
A, X |
|
Analysis of Covariance, Unequal Slopes |
A, X, A*X |
|
Two-Factor Nested Random Effects Model |
A, B[A]&Random |
|
Three-Factor Fully Nested Random Effects Model |
A, B[A]&Random, C[A,B]&Random |
|
Simple Split Plot or Repeated Measures Model |
A, B[A]&Random, C, C*A |
|
Two-Factor Response Surface Model |
X&RS, Z&RS, X*X, X*Z, Z*Z |