Minitab®  Support

Select an alternative model for Discover Best Model (Continuous Response)

Learn more about
Run Predictive Analytics Module > Automated Machine Learning > Discover Best Model (Continuous Response). Click the Select an Alternative Model button after the Model Selection table.
Note

This command is available with the Predictive Analytics Module. Click here for more information about how to activate the module.

When you use Discover Best Model (Continuous Response) to identify the best model type, Minitab Statistical Software produces results for the model with the best value of the accuracy criterion for the analysis, such as the maximum R2. Minitab lets you explore results for other models and other types of models. For example, if another type of model produces similar prediction accuracy, you can determine whether the same predictors are important in each type of model.

The available options depend on the type of model. For multiple regression and CART® models, you can examine the results for the best model from the search. For Random Forests®,TreeNet®, and MARS® models you can examine results from any of the models in the search. For Random Forests®, TreeNet® and MARS® models, you can also tune the hyperparameters to look for combinations that produce even better values than the hyperparameters in the search.

Random Forests®

Select an existing model to produce results for one of the models from the search. Specify hyperparameters to fit new models to look for combinations of hyperparameters that improve the performance of the model.

Select an existing model

In the search for the best type of model, the analysis produces up to 3 Random Forests® models with different minimum sizes for internal nodes. Select a model from the list and click Display Results to produce results for that model.

Specify hyperparameters to fit new models

The analysis requires that you specify the first 3 hyperparameters. The inclusion of a bootstrap sample size less than the training data size is optional. Click Display Results to evaluate the hyperparameters for the new models. The results include a table that compares the optimality criteria for the different combinations of hyperparameters and the model results for the model with the best value of the optimality criterion, such as the maximum R2.
Number of predictors for node splitting
Specify 1 to 3 numbers of predictors to consider for each node split. Typically, the analysis works well when you consider the square root of the total number of predictors. However, some data sets have associations among the predictors that lead to improved model performance when the analysis considers a larger or smaller number of predictors for each node.
Minimum number of cases to split an internal node
Enter 1 to 3 minimum numbers of cases a node can have and still split into more nodes. By default, the original search includes the numbers 2, 5, and 8.
Number of bootstrap samples to grow trees
Enter a value to determine the number of bootstrap samples and the number of trees produced by the analysis. Enter a value between 3 and 3000.
Specify a bootstrap sample size less than the training data size
Select to enter a value that sets the bootstrap sample size. You must enter a value greater than or equal to 5. If you enter a size that is greater than the training data size, Minitab uses a sample size equal to the training data size.

TreeNet®

Select an existing model to produce results for one of the models from the search. Specify hyperparameters to fit new models to look for combinations of hyperparameters that improve the performance of the model.

Select an existing model

In the search for the best type of model, the analysis produces a TreeNet® model for each combination of hyperparameters. Select a model from the list and click Display Results to produce results for that model.

Specify hyperparameters to fit new models

The analysis requires that you specify all of the hyperparameters. Click Display Results to evaluate the hyperparameters for the new models. The results include a table that compares the optimality criteria for the different combinations of hyperparameters and the results for the model with the best value of the accuracy criterion for the analysis, such as the maximum R2.

Learning rate
Enter up to 10 values. Eligible values are from 0.0001 to 1.
Subsample fraction
Enter up to 10 values. Eligible values are greater than 0 and less than or equal to 1.
Maximum terminal nodes per tree and Maximum tree depth
Choose whether to evaluate the Maximum terminal nodes per tree or the Maximum tree depth. Usually, either choice is a reasonable way to identify a useful model, so that the selection depends only on individual preference.
Maximum terminal nodes
Enter up to 3 values. Eligible values are between 2 and 2000. A value of 2 eliminates the investigation of interactions.
Maximum tree depth
Enter up to 3 values. Eligible values are between 2 and 1000 to represent the maximum depth of a tree. The root node corresponds to a depth of 1. In many applications, depths from 4 to 6 give reasonably good models
Number of predictors for node splitting
Enter up to 3 values. Eligible values are between 1 and the total number of predictors. Usually, the analysis works well when you consider the total number of predictors. However, some data sets have associations among the predictors that lead to improved model performance when the analysis considers a smaller number of predictors for each node.
Number of trees
Enter a value between 1 and 5000 to specify the maximum number of trees to build. The default value of 300 usually provides useful results for the evaluation of the hyperparameter values.
If one or more models of interest have a number of trees that is close to the number of trees that you specify, then consider whether to increase the number of trees. If the number of trees is closer to the maximum number, an increase in the number of trees is more likely to improve the performance of the model.

MARS®

Select an existing model to produce results for one of the models from the search. Specify hyperparameters to fit new models to look for combinations of hyperparameters that improve the performance of the model.

Select an existing model

In the search for the best type of model, the analysis produces a MARS® model with each number of basis functions in the search. Select a model from the list and click Display Results to produce results for that model.

Specify parameters to fit new models

Click Display Results to evaluate the parameters for the new models. The results include the model results for the model with the best value of the optimality criterion, such as the maximum R2.
Maximum number of basis functions
The default value of 30 works well in most cases. Consider a larger value when 30 basis functions seems too small for the data. For example, consider a larger value when you believe that more than 30 predictors are important.
If you are uncertain whether 30 is enough, review the initial results. For example, a larger value is more likely to improve the fit of the model if the R-squared value trends upwards as the analysis adds basis functions.
Minimum number of observations between knots
Allow MARS® to choose
The analysis uses sample size and model complexity to automatically select a value. The automatic value works well in most cases.
User specified
A value of 1 indicates that consecutive data points are eligible to be points where the basis function changes. The value of 1 allows the most rapid changes in the model predictions. Use larger values to create smoother models to explore more general relationships. Such smoother models are sometimes less accurate over certain ranges of the data.
Allowed predictor interactions

Allow predictor interactions up to order that you specify. 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.

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 2
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. The following basis functions are an example of an interaction of order 2:
  • BF1 = max(0, X1 − 800)
  • BF2 = max(0, X2 − 50) * BF1

Multiple Regression

Select Results for multiple regression model and click Display Results to produce the results for the best multiple regression model from the search for the best type of model.

CART®

Select Results for CART® model and click Display Results to produce the results for the best CART® model from the search for the best type of model.

Copyright © 2024 Minitab, LLC. All rights Reserved.