Data considerations for Discover Best Model (Continuous Response)


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

To ensure that your results are valid, consider the following guidelines when you collect data, perform the analysis, and interpret your results.

The response variable should be continuous
A continuous variable can be measured and ordered, and has an infinite number of values between any two values. For example, the diameters of a sample of tires is a continuous variable.

The data for the response variable must be numeric values.

If your response variable is binary, use Discover Best Model (Binary Response).

Predictor variables may be continuous or categorical
You can use a combination of continuous or categorical predictors; however, the column lengths for each predictor must be the same length as the response column. Missing values are allowed.
  • All continuous predictors must be numeric.
  • Categorical predictors can be text or numeric values.
A test set is the default when the number of cases > 2000

Minitab uses cross-validation to compare the models when the number of cases is ≤ 2000. When the number of cases is larger than 2000, Minitab uses a test set. When the data set is large, validation with a test set reduces the time to analyze the data. To learn more about the settings for validation techniques in Discover Best Model (Continuous Response), go to Specify the validation method for Discover Best Model (Continuous Response).

The model should provide a good fit to the data

If the model does not fit the data, the results can be misleading. All of the model types include model summary statistics the describe the performance of the model. Use the results from the cross-validation or from the test set to determine if the model predicts the response well. In the output for a regression model, also use the residual plots to verify that parametric assumptions hold.