Regression

A PMML regression model predicts the value of a numeric dependent variable from one or more independent variables.

Model Element

<RegressionModel functionName="regression" ...

Unsupported Features

Regression models with a "normalizationMethod" attribute set to the value "simplemax", "probit", "cloglog" or "loglog" are not supported.

Regression models with integer or float target field are not supported unless <Targets> element is specified with appropriate castInteger attribute.

Regression models with the <MiningSchema> element containing a reference to a <DerivedField> element are not supported.

Model Outputs

By default the target field will be available as an output field - this is a synonym for the predictedValue feature. Float target fields will always be cast to integer.

Supported Model Output Features Description

predictedValue

The numeric dependent variable that we are predicting.

transformedValue

A value generated via a transformation expression applied to the predicted model output.

decision

A value generated via an expression applied to the predicted model output resulting in a categorized value.