非对称单位根检验和非参数方差比

基于三种退势方法较详细研究了方差比检验在非对称单位根检验中的适用性,并通过MC模拟揭示了其检验势性质。结果表明：在不含趋势项的TAR下,两机制TAR数据落在第一机制的比率是影响方差比检验势的重要因素,且比率越高检验势也越高;三机制TAR中落在中间机制的数据比率会影响检验势,随着比率增加检验势呈下降趋势,但程度不大。在含趋势的TAR下,由于趋势项在数据生成过程中具有支配作用,各种检验势会随着趋势设定的不同而不同。数据在不同机制之间的转换概率越高,则ROLS和RDM退势较OLS退势具有明显优势。

Asymmetric Unit Root Tests and Nonparametric Variance Ratio

This paper studies the applicability of variance ratio tests in the asymmetric unit root test in detail based on three detrending methods,and through a set of Monte-Carlo simulations,the paper reveals some properties of those tests.The simulation results show: first,under threshold Autoregressive models without a drift,the data ratio falling within the first regime has a important impact on variance ratio test power in two-regime TAR,and the higher the ratio is,the higher the test power is;the data falling ...