摘要
研究目标:构建一种基于逐步分类加总、分层检验和有限剔除(Castle)的加总偏误IV校正方法。研究方法:构建来自加总模型内部的单一IV和多重IV,并利用美欧农产品贸易细分与加总数据模型对Castle-IV方法的加总偏误校正效果进行了验证。研究发现:通过Castle方法构建的来自加总模型内部的单一IV和多重IV可以获得加总模型参数的一致估计量,实现加总偏误的有效校正。研究创新:本文通过Castle方法构建来自加总模型内部的IV,创新了传统IV的设计思路。研究价值:这一创新性的Castle-IV方法可以实现加总偏误的有效校正。
Research Objectives: The paper aims to construct a Castle-IV method of correcting on aggregation bias.Research Methods: Based on classified aggregation, stratified test and limited elimination, such as Castle process, the paper proposes a single IV and multiple IVs to obtain the consistent estimator and provided methods of testing weak IV,furthermore, which are testified by practical research. Research Findings:The single IV and multiple IVs can obtain the consistent estimator of the aggregation model.Research Innovations:The paper discovers the Castle-IV methods to cope with the aggregation bias.Research Value: The innovation on IV methods can be widely used in correcting the aggregation bias.
引文
[1]Angrist J.D.,Pischke J.S.,2009,Mostly Harmless Econometrics:An Empiricist's Companion [M],Princeton University Press.
[2]Breiman L.,1996,Bagging Predictors [J],Machine Learning,24(2),123~140.
[3]Byrne J.P.,Kontonikas A.,Montagnoli A.,2007,Unit Roots in Inflation and Aggregation Bias [R],University of Glasgow Working Papers,No.2007-07.
[4]Casasnovas V.L.,Aldanondo A.M.,2015,Input Aggregation Bias in Technical Efficiency with Multiple Criteria Analysis [J], Applied Economics Letters,22(6),430~435.
[5]Giray G.,2014, Aggregated and Disaggregated Import Demand in China:An Empirical Study [J],Economic Modelling, 43,1~8.
[6]Hoffman A.,Giovannini E.,2008, Handbook on Constructing Composite Indicators [R],OECD Statistics Working Paper,No.2008.
[7]Imbs J.,Mumtaz H.,Ravn M.O.,Rey.H.,2005,“Aggregation Bias”Does Explain the PPP Puzzle [R],National Bureau of Economic Research Working Paper,No.11607.
[8]Linden M.,Ray D.,2017,Aggregation Bias-Correcting Approach to the Health-Income Relationship:Life Expectancy and GDP per Capita in 148 Countries,1970~2010 [J],Economic Modelling,61,126~136.
[9]Jun I.,2011,Aggregation Bias Found in Korean Demand for Imported Beef [J],Journal of Rural Development,34(2),75~91.
[10]Pesaran M.H.,Pierwe P.G.,Kumar M.S.,1989, Econometric Analysis of Aggregation in the Context of Linear Prediction Models [J], Econometrica,57(4),861~888.
[11]Staiger D.,Stock J.H.,1997,Instrumental Variables Regression with Weak Instrument [J],Econometrica,65(3),557~586.
[12]Sanderson E.,Windmeijer F.,2016,A Weak Instrument F-test in Linear IV Models with Multiple Endogenous Variables [J],Journal of Econometrics,190(2),212~221.
[13]Tenn S.,2006,Avoiding Aggregation Bias in Demand Estimation:A Multivariate Promotional Disaggregation Approach [J],Quantitative Marketing and Economics,4(4),383~405.
[14]Yang L.S.,Chen S.H.Chang C.L.,Tseng Y.H.,2015,Aggregation Problem in DSGE Model[R],2015 International Conference on Behavioral, Economic and Socio-Cultural Computing,123~129.
[15]黄新飞、陈珊珊、李腾:《价格差异、市场分割与边界效应——基于长三角15个城市的实证研究》[J],《经济研究》2014年第12期。
[16]覃琼霞、江涛:《计量模型中的加总偏误与检验方法研究》[J],《数量经济技术经济研究》2011年第5期。
[17]覃琼霞、江涛:《计量模型中的加总偏误识别与检验》[J],《数量经济技术经济研究》2015年第8期。
[18]袁志刚、高虹:《宏观经济学演进:一个新视角》[J],《改革》2016年第11期。