Predictive analytics provides the means to differentiate among customers based on their likely future behavior. The Pega 7 Platform allows you to develop predictions by analyzing the typical behavior of cases across the intervals and categories of predictors. You first analyze cases separately, then in pairs, and finally as a combination of many groups.
Decision Strategy Management embodies the results of the analyses in a mathematical equation referred to as predictive model. This equation enables you to use the information about a case to calculate a value that predicts the behavior of the cases. The specific meaning of the value depends on its type. Provided the behavior is statistically distinct, predictive power and reliability stem from having many different (grouping for numeric predictors) and category sets. Predictive Analytics Director (PAD) analyzes and develops these relationships automatically. If you want to optimize the results, you can view the relationship graphically and adjust the settings in the Data analysis step.
You can generate and maintain the following types of predictive models in PAD: