1分快三彩票

IAER Seminar 2019-2:Xiaoqi ZHANG -- 活动 -- 东北财经大学

导航
首页 - 活动 - IAER Seminar 2019-2:Xiaoqi ZHANG
活动
IAER Seminar 2019-2:Xiaoqi ZHANG

报告题目:Identify Heterogeneous Labor Migration Patterns from Multi-agent Trajectories on the Basis of Network Games

报 告 人:Xiaoqi ZHANG

报告时间: 2019年10月11日(周五)15:30-17:00

报告地点:博学楼2楼东侧I-206(高等经济研究院会议室)

主办单位:高等经济研究院

【报告人简介】

Dr. Xiaoqi Zhang is an assistant research fellow at Chinese Academy  of Social Sciences and National Development Institute of Southeast University.  He received his Ph. D. from The state University of New York at Buffalo in 2017.  His research focuses on Financial Econometric Analysis (Survival Analysis and  its application in Default Prediction and Insurance Actuarial), Mathematical  Statistics, Theory and Application of Continuous Time Stochastic Processes, Game  Econometrics, Spatial Econometric Method and its application, Topological Data  Analysis and its application in Network Data.

【内容摘要】

The migration intensity among regions could differ significantly for  different agent groups, and this difference is rooted deeply in the  heterogeneity on the agent level. The traditional reconstruction method of  migration network does not pay enough attention to the agent-level  heterogeneity, which leads to bias and low forecsatability. On the other hand,  the migration trajectories in many settings, such as the labor force migration  trajectories and the patient transmission trajectories, are essentially the  consequence of the interaction among all agents in the system. The inference of  the migration network from trajectory data should be able to shed insight into  the underlying interaction mechanisms, while most existing methods fail to do  so. To fill the gap, the current study aims at developing a game-based  constrained maximum likelihood estimation procedure. The procedure can extract  the migration network from the agent-level trajectory data, the network  adjacency matrix can vary from agents to agents so as to fully account for the  agent-level heterogeneity, and the game-theoretical model involved gives a  sketch of the interaction among agents. To accommodate our approach to massive  trajectory data, we develop an approximation trick which converts the  agent-level migration probabilities to a set of functions that can be explicitly  expressed and only have the parameters and agent-level features as arguments.  Consequently, it is only needed to search within a parameter space with fixed  dimension for solving the maximum likelihood estimator. The termination time of  the algorithm won't diverge along with the increasing number of trajectories. We  demonstrate the effectiveness of our approach on a Chinese labor force migration  dataset with 80,000 job-seekers. The cross validation is used to test the  forecast accuracy of the approach. The result is robust and supports that our  approach presents a useful paradigm for understanding the complex relationship  among migration patterns, age.

详情参见高等经济研究院网站:http://iaer.dufe.edu.cn/


撰稿:王杰 审核:齐鹰飞 单位:高等经济研究院

新 闻