题 目：Spatial Modeling Approach for Dynamic Network Formation and Interactions

报告人：韩晓�  厦门大学王亚南经济研究院  副教授

主持人：梁友莎  讲师  浙江大学经济学院

时 间：2018年5月18日（周五）  14:00―15:30

地 点：浙江大学玉泉校区经济学院418室  

Abstract

This study primarily seeks to answer the following question: How do social networks evolve over time and affect individual economic activity? To provide an adequate empirical tool to answer this question, we propose a new modeling approach for longitudinal data of networks and activity outcomes. The key features of our model are the inclusion of dynamic features and the use of latent variables to determine unobserved individual traits in network formation and activity interactions. In particular, these unobserved traits vary over time. 

The proposed model combines two well-known models in the field: latent space model for dynamic network formation and spatial dynamic panel data model for network interactions. This combination reflects real situations, where network links and activity outcomes are interdependent and jointly influenced by unobserved individual traits. Moreover, this combination enables us to (1) manage the endogenous selection issue inherited in network interaction studies, and (2) investigate the effect of homophily and individual heterogeneity from observed and unobserved characteristics in the network formation. We develop a tractable Bayesian Markov chain Monte Carlo approach to estimate the model. We also provide a Monte Carlo experiment to analyze the performance of our estimation method and apply the model to a longitudinal student network data in Taiwan to study the friendship network formation of students and peer effect on academic performance.

报告人简介：

 韩晓�在浙江大学获得经济学学士学位，俄亥俄州立大学获得经济学博士学位。现任厦门大学王亚南经济研究院副教授。

CRPE秘书处

2018-5-15