Best Subsets Regressionoverview how to example data see also
Best Subsets RegressionStat > Regression > Regression > Best Subsets
Best subsets regression identifies the subset models that produce the
highest R
values from full set of the predictor variables
that you specify. Best subsets regression is an efficient way to identify
models that achieve your goals with as few predictors as possible. Subset
models may actually estimate the regression coefficients
and predict future responses with smaller variance than the full model
using all predictors [22].
Minitab examines all possible subsets of the predictors, beginning with all models containing one predictor, and then all models containing two predictors, and so on. By default, Minitab displays the two best models for each number of predictors.
For example, suppose you conduct a best subsets regression with three predictors. Minitab will report the best and second best one-predictor models, followed by the best and second best two-predictor models, followed by the full model containing all three predictors.
Response: Enter the column containing the response (Y) variable.
Free predictors: Enter the columns containing the candidate predictor (X) variables. You can specify up to 31 variables, though large models require long computation time.
Predictors in all models: Select columns containing variables you want included as predictors in every model. Columns entered here must not be listed in Free predictors. If you are analyzing a large data set with more than 15 predictors, consider including certain predictors here in order to decrease the number of free variables and speed up the computations. The maximum number of variables which can be entered is equal to 100 minus the number of variables entered in Free predictors.