时间：2018年4月11（星期三）13：30-15：30

地点：学院南路校区，学术会堂604

报告一题目：国内产业转移还是国际产业转移——基于世界投入产出表的研究

报告人：中国人民大学经济学院、林晨副教授

摘要：本文采用世界投入产出数据库和中国各省投入产出表，从分配的视角探讨我国东部地区产业转移的方向。分析结果显示我国省份之间资本劳动收益比的巨大差异令部分中西部省份相对于印度尼西亚等发展中国家具有承接劳动密集型产业的比较优势。资本劳动收益比的差异源自于限制资本劳动自由流动的制度和非制度性安排。若限制资本劳动自由流动的障碍源自于地方保护主义，则因此导致的国内产业转移并非是有效率的。

报告人简介：林晨现任中国人民大学经济学院副教授。2010年于日本早稻田大学获经济学学博士学位，研究方向为投入产出经济学、数理经济学等。

报告二题目：Estimation of Panel Data Models for US State Level House Price with Many Instruments

报告人：yl7703永利官网、黄白博士

摘要：When the regressors are endogenous due to simultaneity or measurement errors, the fixed effect (FE) estimator for a panel data model is inconsistent. We extend the FE estimator using instrumental variables (denoted FE-2SLS) to analyze the US house prices using state level panel data. However, we find that the FE-2SLS estimator is sensitive to the number of selected instruments and is inconsistent when the number increases. We show that using regularization methods such as Lasso and SCAD for the selection of instruments would make the FE-2SLS estimator more robust and restore its consistency when there are many instruments. Also, extending Hansen (2017) to the structural panel data models, we consider a combined estimator of the FE and FE-2SLS estimators and provide its asymptotic properties. We show that the combined estimator has the asymptotic risk strictly smaller than that of the FE-2SLS estimator when the FE-2SLS is consistent. Our Monte Carlo analysis shows the asymptotic theory carries over to finite sample only when the small number of good instruments are carefully selected, while otherwise the combined estimator can be worse than the FE-2SLS estimator from using too many instruments even when endogeneity is large. Guarded with these theoretical and numerical ndings, a careful empirical study is conducted for the economics of real house price using state level panel data.

报告人简介：黄白博士2017年于University of California, Riverside取得计量经济学方向博士学位，师从Aman Ullah杰出教授（Nonparametric, Panel Data, Finite sample方向专家）和Tae-Hwy Lee教授（时间序列数据，模型平均，高维数据，大数据方向专家）。主要的研究兴趣包括面板数据，区间数据，非参数估计和高维数据等。