Spotfire Statistica® Stepwise Model Builder

This is a "what-if" tool. Stepwise Model Builder's goal is to facilitate the identification of models based on predictors chosen by the user at each step. At each step, the program will compute various predictor statistics for predictors in the current model, and predictors (predictor candidates) not in the current equation. Statistics reflecting on the overall model quality are also computed.

Users can build models by manually selecting the most important predictors into the regression equation one step at a time, using criteria of statistical significance for the prediction as well as policy and other criteria. By moving selected variables or groups of variables into the prediction and equation, and removing others from that equation, what-if (scenario) analyses are possible to assess the impact of certain model assumptions, policy, or regulatory constraints (e.g., on predictors that are not permitted). Thus, analysts can build models that are parsimonious, consistent with policies, guidelines, and regulatory constraints but are also as accurate as possible.

Stepwise model builder is available for the following techniques for Spotfire Statistica® Basic Academic, Spotfire Statistica® Desktop and Spotfire Statistica® Analyst products. 

• Cox's proportional hazards: distribution-free model in which predictors are related to lifetime multiplicatively
• Linear regression: learn more about the relationship between several independent or predictor variables and a dependent or criterion variable
• Logistic Regression: model binary outcome variables - such as credit default, insurance or warranty claim incidences

Stepwise model builder is available for the following techniques for Spotfire Statistica® Modeler, Spotfire Statistica® Data Scientist, Spotfire Statistica® Comprehensive and Spotfire Statistica® Ultimate Academic products.

• Boosted tree; creates boosting classification trees for continuous and categorical predictors
• Cox's proportional hazards: distribution-free model in which predictors are related to lifetime multiplicatively
• Linear regression: learn more about the relationship between several independent or predictor variables and a dependent or criterion variable
• Logistic Regression: model binary outcome variables - such as credit default, insurance or warranty claim incidences
• Random Forest; regression-type problems (to predict a continuous dependent variable) as well as classification problems (to predict a categorical dependent variable)