Select the interactions to consider in the model. Partial dependence plots are not
available for an analysis that considers interactions.
An interaction means that the effect of a predictor depends on the value of other
predictors. For example, the rate at which grain dries in an oven depends on the time in
the oven, but the effect of time depends on the temperature of the oven. The time and
temperature variables interact.
The limitation of interactions still allows multiple basis functions for single
variables. To force a continuous predictor to have only a linear effect, go to the
Options for the analysis and disable transformations for the predictor.
- Do not allow
any interactions (Additive model)
: Do not allow predictor
interactions. In this case, Minitab uses the additive model where the basis
functions do not interact.
- Allow all interactions up to order: Order specifies the number of different predictors that can be in a basis
function. For example, an order of 2 indicates that the effect of a predictor
can depend on the value of 1 other predictor. An order of 3 indicates that the
effect of a predictor can depend on the value of 2 other predictors. The
following basis functions are an example of an interaction of order 3:
- BF1 = max(0, X1 − 800)
- BF2 = max(0, X2 − 50) * BF1
- BF3 = max(0, X3 − 10) * BF 2
An order of 4 indicates that the effect of a predictor can depend on the
value of 3 other predictors.
- Select specific predictor interactions up to
order: Order specifies the number of
different predictors that can be in a basis function. For example, an order of 2
indicates the effect of a predictor can depend on the value of 1 other
predictor. An order of 3 indicates that the effect of a predictor can depend on
the value of 2 other predictors. An order of 4 indicates that the effect of a
predictor can depend on the value of 3 other predictors. The following basis
functions are an example of an interaction of order 3:
- BF1 = max(0, X1 − 800)
- BF2 = max(0, X2 − 50) * BF1
- BF3 = max(0, X3 − 10) * BF 2
-
In Predictors, enter the columns that
contain the predictors that you want to allow for the interactions. Enter at
least as many predictors as the order of the interactions. For example, to
consider interactions of order 2, enter 2 or more predictors. If you specify
no predictors, then the analysis considers interactions among all of the
predictors.
