Drawing on recent "event history" analytical methods from biostatistics, engineering, and sociology, this clear and comprehensive monograph explains how longitudinal data can be used to study the causes of deaths, crimes, wars, and many other human events. Allison shows why ordinary multiple regression is not suited to analyze event history data, and demonstrates how innovative regression - like methods can overcome this problem. He then discusses the particular new methods that social scientists should find useful.

This book is an accessible, practical and comprehensive guide for researchers from multiple disciplines including biomedical, epidemiology, engineering and the social sciences. Written for accessibility, this book will appeal to students and researchers who want to understand the basics of survival and event history analysis and apply these methods without getting entangled in mathematical and theoretical technicalities. Inside, readers are offered a blueprint for their entire research project from data preparation to model selection and diagnostics. Engaging, easy to read, functional and packed with enlightening examples, ‘hands-on’ exercises, conversations with key scholars and resources for both students and instructors, this text allows researchers to quickly master advanced statistical techniques. It is written from the perspective of the ‘user’, making it suitable as both a self-learning tool and graduate-level textbook. Also included are up-to-date innovations in the field, including advancements in the assessment of model fit, unobserved heterogeneity, recurrent events and multilevel event history models. Practical instructions are also included for using the statistical programs of R, STATA and SPSS, enabling readers to replicate the examples described in the text.

Event history analysis has been a useful method in the social sciences for studying the processes of social change. However, a main difficulty in using this technique is to observe all relevant explanatory variables without missing any variables. This book presents a general approach to missing data problems in event history analysis which is based on the similarities between log-linear models, hazard models and event history models. It begins with a discussion of log-rate models, modified path models and methods for obtaining maximum likelihood estimates of the parameters of log-linear models. The author then shows how to incorporate variables with missing information in log-linear models - including latent class models, m

The aim of this book is to bridge the gap between standard textbook models and a range of models where the dynamic structure of the data manifests itself fully. The common denominator of such models is stochastic processes. The authors show how counting processes, martingales, and stochastic integrals fit very nicely with censored data. Beginning with standard analyses such as Kaplan-Meier plots and Cox regression, the presentation progresses to the additive hazard model and recurrent event data. Stochastic processes are also used as natural models for individual frailty; they allow sensible interpretations of a number of surprising artifacts seen in population data. The stochastic process framework is naturally connected to causality. The authors show how dynamic path analyses can incorporate many modern causality ideas in a framework that takes the time aspect seriously. To make the material accessible to the reader, a large number of practical examples, mainly from medicine, are developed in detail. Stochastic processes are introduced in an intuitive and non-technical manner. The book is aimed at investigators who use event history methods and want a better understanding of the statistical concepts. It is suitable as a textbook for graduate courses in statistics and biostatistics.

Social scientists are interested in events and their causes. Although event histories are ideal for studying the causes of events, they typically possess two features—censoring and time-varying explanatory variables—that create major problems for standard statistical procedures. Several innovative approaches have been developed to accommodate these two peculiarities of event history data. This volume surveys these methods, concentrating on the approaches that are most useful to the social sciences. In particular, Paul D. Allison focuses on regression methods in which the occurrence of events is dependent on one or more explanatory variables. He gives attention to the statistical models that form the basis of event history analysis, and also to practical concerns such as data management, cost, and useful computer software. The Second Edition is part of SAGE’s Quantitative Applications in the Social Sciences (QASS) series, which continues to serve countless students, instructors, and researchers in learning the most cutting-edge quantitative techniques.