Bayesian analysis of event history models with unobserved heterogeneity via markov chain Monte Carlo - Application to the explanation of fertility decline

This article describes an interesting application of Markov chain Monte Car lo (MCMC). MCMC is used to assess competing explanations of marital fertili ty decline. Data collected during the World Fertility Study in Iran are ana lyzed using methods developed to perform discrete time event history analys es in which unobserved heterogeneity is explicitly accounted for. The usual age-period-cohort identifiability problem is compounded by the presence of a fourth clock, duration since previous birth, and a fifth clocklike varia ble, mother's parity. The authors resolve this problem by modeling some of the clocks parametrically using codings suggested by alternating conditiona l expectation (ACE) and Bayes factors to decide which clocks are necessary. Compound Laplace-Metropolis estimates are used to compute Bayes factors fo r comparing alternative models. The new methods enable the authors to concl ude that Iran's fertility decline was primarily a period effect and not a c ohort effect, that it started before the Family Planning Program was initia ted, that it was the same for women at all educational levels but varied de pending on husband's education, and that it was greatest in the largest cit ies, particularly Tehran.