Bayesian Cohort Models for General Cohort Table Analyses

摘要: New Bayesian cohort models designed to resolve the identification problem in analysis are proposed this paper. At first, basic model which represents statistical structure of time-series social survey data terms age, period and effects is explained. The logit for qualitative from a binomial distribution normal-type quantitative normal considered as two special cases model. In order overcome analysis, approach adopted, based on assumption that effect parameters change gradually. A information criterion ABIC introduced selection optimal This so flexible both can be made applicable, not only standard tables but also general range age group equal interval between periods. practical utility demonstrated by analysing sets literature analysis.