太阳成集团tyc411(中国)有限公司-百度百科九十周年院庆系列活动之十五

概率统计学术报告（6月16-17 日Liang Peng 教授）

报告人：Liang Peng 教授， Georgia State University， 美国

题目：Inference for Mortality Model and Predictive Regression（I）

时间：2018年6月16日下午2:00---5:00

地点：浙江大学紫金港校区启真酒店4楼梨洲厅

摘要：The first part of this talk is about mortality models in actuarial science. Although the Lee-Carter model has become a benchmark in modeling mortality rates and forecasting mortality risk, there exist some serious issues on its inference and  interpretation in the literature of actuarial science. After pointing out these pitfalls and  misunderstandings, we propose a modified Lee-Carter model, provide a sound statistical inference and  derive the asymptotic distributions of the proposed estimators and unit root test when the mortality index is nearly integrated and errors in the model satisfy some mixing  conditions. After a unit root hypothesis is not rejected,    future mortality forecasts can be obtained via the proposed inference. An application of the proposed unit root test to   US mortality rates rejects the unit root hypothesis for the female and combined mortality rates, but does not reject the unit root hypothesis for the male mortality rates.

欢迎大家参加!

太阳成集团tyc411(中国)有限公司-百度百科九十周年院庆系列活动之十六

概率统计学术报告（Liang Peng 教授）

报告人：Liang Peng 教授， Georgia State University， 美国

题目：Inference for Mortality Model and Predictive Regression（II）

时间：2018年6月17日下午2:00---5:00

地点：浙江大学紫金港校区启真酒店4楼梨洲厅

摘要：The second part of this talk is about predictive regression in financial econometrics. Testing for predictability of asset returns has been a long history in economics and finance. Recently, based on a simple predictive regression, Kostakis, Magdalinos and Stamatogiannis (2015) derived a Wald type test based on the context of the extended instrumental variable (IVX) methodology for testing predictability of stock returns and Demetrescu (2014) showed that the local power of the standard IVX-based test could  be improved in some cases when a lagged predicted variable is added to the predictive regression on purpose, which poses a general important question on whether a lagged predicted variable should be included in the model or not. We propose novel robust   procedures for testing both the existence of a lagged predicted variable and the predictability of asset returns in a predictive regression regardless of regressors being stationary or nearly integrated or unit root. A simulation study confirms the good finite sample performance of the proposed tests before applying the proposed tests to some real datasets in finance.

欢迎大家参加!

联系人：苏中根，suzhonggen@zju.edu.cn

浙江大学统计研究所

 2018年6月3日