报告人简介：

Xufeng Niu教授在芝加哥大学获得统计学博士学位后，一直在佛罗里达州立大学统计系工作至今。研究兴趣主要有：Time series analysis、Linear and non-linear models、Spatial statistics、Environmental data analysis and asymptotic theory、Longitudinal data analysis，在Journal of the Acoustical Society of America (JASA)、Economic Modelling、Int J Environ Res Public Health等学术刊物发表论文多篇。

 报告简介：

Motivated by understanding the devastating financial crisis in 2008 that was partially caused by underestimation of financial risk, we propose a class of time-varying mixture models for risk analysis and management. Previously one commonly used approach to estimate VaR is the Variance-Covariance method, in which normal distribution is usually assumed for asset returns that may underestimate the real risk. To address this issue, we will use time-varying mixture models of distributions from different families for the estimation of VaR, in which weights for each distribution at different time periods will be estimated based on Macroeconomic indices. In addition, time series models will be used to examine the correlation structure of the time-varying weights. Statistical properties of the models will be investigated, including estimation methods and asymptotic results of the parameter estimates. Simulation study will be carried out to evaluate different estimation methods and model selection procedures.