Publication date: 04/12/2021

Image shown hereThe Model Screening Report

The content of the Model Screening report depends on both the type of crossvalidation implemented and the modeling type of the response variable. At the top of the Model Screening report, the name of the data table and column name of the response variable are shown. If a validation column is specified, the name of that column is also shown.

By default, the Details report appears next. This report contains the individual model reports for all of the models for each fold and trial. These reports are the actual output from each platform and contain all reports and command features. For example, if you decide on a model that is best, you can go to the corresponding platform report in the Details report and save the desired prediction formula column. You can hide the Details report by selecting the Remove Live Reports option in the launch window.

If either the K Fold Crossvalidation option or the Nested Crossvalidation option is specified in the launch window, the Summary Across the Folds report appears. This report is a summary of the measures of fit averaged across the test folds.

Measures of fit are shown for each individual model fit, for the training, validation, and test sets if applicable. If crossvalidation is used by specifying a validation column in the launch window, only the individual model fit measures of fit are shown in the report.

Want more information? Have questions? Get answers in the JMP User Community (