Discrete Dependent Variables and Duration Models

  • Robert C. Blattberg
  • Byung-Do Kim
  • Scott A. Neslin
Part of the International Series in Quantitative Marketing book series (ISQM, volume 18)


Probably the most common statistical technique in predictive modeling is the binary response, or logistic regression, model. This model is designed to predict either/or behavior such as “Will the customer buy?” or “Will the customer churn?” We discuss logistic regression and other discrete models such as discriminant analysis, multinomial logit, and count data methods. Duration models, the second part of this chapter, model the timing for an event to occur. One form of duration model, the hazard model, is particularly important because it can be used to predict how long the customer will remain as a current customer. It can also predict how long it will take before the customer decides to make another purchase, switch to an upgrade, etc. We discuss hazard models in depth.


Hazard Rate Multinomial Logit Model Default Probability Duration Model Linear Probability Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  • Robert C. Blattberg
    • 1
    • 2
  • Byung-Do Kim
    • 3
  • Scott A. Neslin
    • 4
  1. 1.Kellogg School of ManagementNorthwestern UniversityEvanstonUSA
  2. 2.Tepper School of BusinessCarnegie-Mellon UniversityPittsburghUSA
  3. 3.Graduate School of BusinessSeoul National UniversitySeoulKorea
  4. 4.Tuck School of BusinessDartmouth CollegeHanoverUSA

Personalised recommendations