摘要
近年来,门槛模型在实证研究中得到了广泛的应用,其中门槛存在性检验常常是实证应用中关注的重点。然而,因受Davies问题的影响,门槛存在性检验必须借助Bootstrap来实现,但迄今为止,文献很少对不同Bootstrap检验流程的有限样本性质展开系统性比较和分析,鉴于此,文章在典型数据生成过程下结合蒙特卡洛模拟对文献中常见的Bootstrap检验流程的检验功效、检验水平等问题展开比较研究,然后基于模拟分析的结论,提出在实际应用时选择Bootstrap检验流程的若干建议。
In recent years, threshold model has been widely used in empirical research, in which the threshold existence test is often the focus of empirical application. However, due to the influence of the Davies problem, the threshold existence test must be realized by means of Bootstrap, but so far few literatures have systematically compared and analyzed the limited sample properties of different Bootstrap test procedures. In view of this, combined with Monte Carlo simulation under the typical data generation process, this paper makes a comparative study on the test efficacy and level of Bootstrap test process commonly seen in literature.And then according to the results of simulation analysis, the paper presents some suggestions for the selection of Bootstrap inspection process in practical application.
引文
[1]Hansen B E.Inference When a Nuisance Parameter Is Not Identified Under the Null Hypothesis[J].Econometrica,1996.
[2]Hansen B E.Sample Splitting and Threshold Estimation[J].Econometrica,2000,68(3).
[3]Hansen B E.Threshold Effects in Non-dynamic Panels:Estimation,Testing,and Inference[J].Journal of Econometrics,1999,93(2).
[4]Balke N S,Fomby T B.Threshold Cointegration[J].International Economic Review,1997,38(3).
[5]Gonzalo J,Pitarakis J Y.Threshold Effects in Cointegrating Relationships[J].Oxford Bulletin of Economics and Statistics,2006,(68).
[6]欧阳志刚.协整向量、调节参数同为非线性的阈值协整检验[J].数量经济技术经济研究,2014,(9).
[7]Davies R B.Hypothesis Testing When a Nuisance Parameter Is Present Only Under the Alternative[J].Biometrika,1977,64(2).
[8]Davies R B.Hypothesis Testing When a Nuisance Parameter Is Present Only Under the Alternative[J].Biometrika,1987,74(1).
[9]Andrews D W K,Ploberger W.Optimal Tests When a Nuisance Parameter is Present Only Under the Alternative[J].Econometrica,1994.
[10]Hansen B E.Regression Kink With an Unknown Threshold[J].Journal of Business&Economic Statistics,2017,35(2).
[11]Porter J,Yu P.Regression Discontinuity Designs With Unknown Discontinuity Points:Testing and Estimation[J].Journal of Econometrics,2015,189(1).
[12]Hodge A,Shankar S.Single-Variable Threshold Effects in Ordered Response Models With an Application to Estimating the Income-Hap iness Gradient[J].Journal of Business&Economic Statistics,2017,34(1).
[13]Davidson J,Monticini A,Peel D.Implementing the Wild Bootstrap Using a Two-point Distribution[J].Economics Letters,2007,96(3).