Abstract
Abstract Longitudinal studies are an increasingly popular method of data collection in a wide range of disciplines, including medicine, public health, sociology, economics, and demography. Many of these studies produce data in the form of event histories (Tuma and Hannan 1978), where individuals are followed over a period of time and dates of events are recorded; for example, births or marital changes in a demographic survey, progression of disease in a clinical trial, changes of employment status in an income survey, or pollution episodes in an environmental study. It is now widely recognized that the analysis of data in this form requires the methodology of survival analysis (see e.g. Kalbfleisch and Prentice 1980), and extensions of that methodology to handle recurring events. This chapter discusses problems arising from incomplete data in the analysis of event histories.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
