Advertisement

Regression Models for Binary Data

  • Jeffrey S. Simonoff
Chapter
  • 1.2k Downloads
Part of the Springer Texts in Statistics book series (STS)

Abstract

The loglinear models of the previous four chapters are designed for count data, where a Poisson or multinomial distribution is appropriate. The most basic form of categorical data, however, is binary — 0 or 1. It is often of great interest to try to model the probability of success (the outcome coded 1) or failure (the outcome coded 0) as a function of other predictors. Consider a study designed to investigate risk factors for cancer. Attributes of people are recorded, including age, gender, packs of cigarettes smoked, and so on. The target variable is whether or not the person has lung cancer (a 0/1 variable, with 0 for no lung cancer and 1 for the presence of lung cancer) . A natural question is then “What factors can be used to predict whether or not a person will have lung cancer?” Substitute businesses for people, financial characteristics for medical risk factors, and whether a company went bankrupt for whether a person has cancer, and this becomes an investigation of the question “What financial characteristics can be used to predict whether or not a business will go bankrupt?” The models of this chapter are designed to address such questions.

Keywords

Logistic Regression Model Logistic Regression Binary Data Wald Statistic Covariate Pattern 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media New York 2003

Authors and Affiliations

  • Jeffrey S. Simonoff
    • 1
  1. 1.Leonard N. Stern School of BusinessNew York UniversityNew YorkUSA

Personalised recommendations