Regression Models for Categorical and Limited Dependent Variables

Autorentext

Scott Long is Distinguished Professor and Chancellor s Professor of Sociology and Statistics at Indiana University, Bloomington. He teaches quantitative methods both at Indiana University and at the ICSPR Summer Program. His earlier research examined gender differences in the scientific career. In recent years, he has collaborated with Eliza Pavalko, Bernice Pescsolido, John Bancroft, Julia Heiman and others in studies of health and aging, stigma and mental health, and human sexuality.

Klappentext

A unified treatment of the most useful models for categorical and limited dependent variables (CLDVs) is provided in this book. Throughout, the links among the models are made explicit, and common methods of derivation, interpretation and testing are applied. In addition, the author explains how models relate to linear regression models whenever possible.

After a review of the linear regression model and an introduction to maximum likelihood estimation, the book then: covers the logit and probit models for binary outcomes; reviews standard statistical tests associated with maximum likelihood estimation; and considers a variety of measures for assessing the fit of a model. J Scott Long also: extends the binary logit and probit models to ordered outcomes; presents the multinomial and conditioned logit models for nominal outcomes; considers models with censored and truncated dependent variables with a focus on the tobit model; describes models for sample selection bias; presents models for count outcomes by beginning with the Poisson regression model; and compares the models from earlier chapters, discussing the links between these models and others not discussed in the book.

Inhalt

Introduction
Continuous Outcomes
Binary Outcomes
Testing and Fit
Ordinal Outcomes
Nominal Outcomes
Limited Outcomes
Count Outcomes
Conclusions