学术报告
报告题目：Robust State-Space Model For Fish Stock
Maturities
报 告 人：Dr. Ximing Xu
Department of Mathematics and Statistics
Dalhousie University, Canada
时 间：2014年12月24日(星期三)上午10:00
地 点：数学楼第三报告厅
摘 要：We propose a robust state-space model to the 3Ps cod maturity data in a DFO's annual sampling program, where the density functions of the observations are Binomial, while the latent state processes are modeled as AR(1). The maximum likelihood estimators of the unknown parameters are obtained based on Laplace approximation to the marginal likelihood of the observations. The sequential particle filtering and smoothing techniques are then applied to recover the underlying dynamics of the maturities through time. Our approach is compared with the sequential Monte Carlo method with state augmentation both in simulations and in the cod maturity data. Issues on the robustness of our procedure are also discussed.
欢迎广大师生参加！
统计研究院
2014年12月22日