方法对比

	Retirement Transition Event-History Analysis	Multistate Life Table
领域	Social Gerontology	人口学
方法族	Survival analysis	Survival analysis
起源年份	1998	1975
提出者	Mark D. Hayward and colleagues	Andrei Rogers, Robert Schoen and collaborators
类型	Time-to-event hazard model of labor-force exit	Nonparametric life table with multiple living states and transitions
开创性文献	Hayward, M. D., Friedman, S., & Chen, H. (1998). Career trajectories and older men's retirement. The Journals of Gerontology Series B, 53B(2), S91-S103.	Preston, S. H., Heuveline, P., & Guillot, M. (2001). Demography: Measuring and Modeling Population Processes. Blackwell. ISBN: 9781557864512
别名	Retirement Hazard Model, Labor-Force Exit Survival Analysis, Retirement Timing Event-History Model, Discrete-Time Retirement Model	Increment-Decrement Life Table, Multiple-State Life Table, Multistate Demography, Çok Durumlu Yaşam Tablosu
相关	3	4
摘要	Retirement transition event-history analysis applies survival and hazard modeling to the timing of the move out of the labor force in later life, treating retirement as a datable event whose risk unfolds over time. Rather than asking only whether someone is retired, it models the rate at which still-working older people retire at each age or duration, and how that rate depends on health, pensions, career history, and other life-course factors. Mark Hayward and colleagues' 1998 study of older men's retirement exemplifies the approach, showing that occupational and career trajectories shape the timing of labor-force exit, with different career conditions mattering at different stages. The method handles the central problem that many people are still working when observed, through right-censoring, and it accommodates covariates that change over time such as deteriorating health or pension eligibility. It can be implemented as a continuous-time proportional-hazards model or as a discrete-time model on person-period data. The result is a life-course account of why people retire when they do, expressed as transition rates and hazard ratios.	The multistate life table, also called the increment-decrement life table, generalizes the ordinary life table to populations that move among several living states — such as healthy and disabled, married and unmarried, or employed and unemployed — as well as the absorbing state of death. Using age-specific transition rates organized in matrices, it tracks the flows of a synthetic cohort among states and yields state-specific expectancies, such as the years a person can expect to spend healthy versus disabled.