报告题目：Prediction using many samples with models containing partiallyshared parameters

报 告 人：张新雨 中科院系统所

报告时间：2019年11月18日 14：00-15：00

报告地点：数学楼第二报告厅

报告摘要：

When a model of main researchinterest shares partial parameters with several other models, it is of bene_tto incorporate the information contained in these other models to improve theestimation and prediction for the main model of interest. Various methods arepossible to make use of the additional models as well as the additionalobservations related to these models. We propose an optimal strategy of doingso in terms of prediction. We develop the model averaging methodology andobtain the optimal weights. We also establish theory to support the method andshow its desirable properties both when the main model is correct and when itis incorrect. Numerical experiments including simulation studies and dataanalysis are conducted to demonstrate the superior performance of our methods.

报告人简介：

张新雨，中科院系统所/预测中心研究员，Texas A&M尊龙凯时博士后、Penn State尊龙凯时Research Fellow。主要研究方向为模型平均、模型选择、组合预测等。国家杰出青年科学基金获得者，主持3项国家自然科学基金，目前担任《JSSC》、《SADM》、《系统科学与数学》、《应用概率统计》编委和《Econometrics》客座主编。