Predictive Analytics


Predictive analytics (sometimes referred to as predictive modeling) involves the use of current and historical data to make predictions about future trends, behaviors, and events. Typically, statistics and data mining techniques are applied to locate and exploit patterns in the data.

Predictive modeling involves the identification of specific variables that are correlated to a target event or behavior. These variables contribute to the formation of a mathematical formula used to attach probabilities or odds (score), to each record (customer or prospect), representing the likelihood of the target event/behavior occurring. A list of records, sorted by score, can be used for targeted marketing, sales, advertising, assessing credit-worthiness, determining propensity to buy specific products, likelihood to churn, and many other things. Predictive modeling is regularly used in retail, telecommunications, insurance, healthcare, medicine, science, pharmaceuticals, and other areas.

Predictive analytics also refers to additional disciplines such as descriptive and decision modeling, and optimization that can be used in business to help make decisions, generate segmentation schemes, and optimize processes.


Common Uses for Predictive Analytics

Common Approaches and Technologies

  • Forecasting
  • Neural Networks
  • Genetic Algorithms
  • Decision Trees (Classification)
  • Clustering
  • Logistic Regression
  • Survival Analysis

learn more