摘要：Unlike majority of the researches which focus on the unemployment rate, this paperstudies the number of claims submitted for jobless compensations. Using nonhomogeneousPoisson process, this study constructs Bayesian stochastic analysis models to describe thebehaviors of unemployment claims submitted to Washington DC government. Applying themodels towards real claim data with implementations of the Markov chain Monte Carlo(MCMC) method, the posterior inferences generated reveals that the model of nonhomogeneousPoisson process with random effects performs better in fitting the data and predicting for futuretime periods. The purposes of the study are to identify influential socioeconomic factorsaffecting the claims and to provide precise predictions of the number of claims for the policymakers to improve the allocation of financial resources and carry out effective programs forhelping unemployed DC residents in financial hardship to get and retain their jobs, and stabilizelocal labor force as well as enhance the social stability.
关键词：unemployment claims; nonhomogeneous Poisson; random effects; MCMC; Bayesian;inferences
