Specify the default settings for stepwise regression

Tools > Options > Linear Models > Stepwise

Specify the stepwise regression method, parameters, number of alternatives to show, and the display of PRESS and predicted R2. The changes you make to the defaults remain until you change them again, even after you exit Minitab. Minitab uses the settings you specify here each time you use one of the following analyses:
  • Stat > Regression > Regression > Fit Regression Model
  • Stat > Regression > Binary Logistic Regression > Fit Binary Logistic Model
  • Stat > Regression > Poisson Regression > Fit Poisson Model
  • Stat > ANOVA > General Linear Model > Fit General Linear Model
  • Stat > DOE > Screening > Create Screening Design
  • Stat > DOE > Factorial > Analyze Factorial Design
  • Stat > DOE > Factorial > Analyze Variability
  • Stat > DOE > Response Surface > Analyze Response Surface Design
  • Stat > DOE > Mixture > Analyze Mixture Design
Default method
  • None: Fit with the model with all the terms that you specify in Terms.
  • Stepwise: Start with an empty model, then add or remove a term for each step. You can specify an initial set of terms to include in the model.
  • Forward selection: Start with an empty model, then add the most significant potential term to the model for each step.
  • Backward elimination: Start with all potential terms in the model, then remove the least significant term for each step.
Stepwise Parameters
  • Alpha to enter: Specify the criterion for entering a new term in the model.
  • Alpha to remove: Specify the criterion for removing term from the model.
Forward Selection Parameter
  • Alpha to enter: Specify the criterion for entering a new term in the model.
Backward Elimination Parameter
  • Alpha to remove: Specify the criterion for removing term from the model.
Display the table of model selection details
Display information about the stepwise procedure in the Session window.
  • Details about the method: Display the type of stepwise procedure and the alpha values that are used to add and/or remove a term from the model.
  • Include details for each step: Display the coefficients, p-values, Mallows' Cp, and model summary statistics for each step of the procedure.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy