协整平滑转移回归中的线性检验             ——基于完全修正最小二乘法的扩展

统计研究 ›› 2010, Vol. 27 ›› Issue (3): 70-75.

协整平滑转移回归中的线性检验 ——基于完全修正最小二乘法的扩展

欧阳志刚

收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-03-15 发布日期:2010-03-15

Testing Linearity in Cointegrating Smooth Transition Regressions ——Based on Fully Modified OLS

Ouyang Zhigang

Received:1900-01-01 Revised:1900-01-01 Online:2010-03-15 Published:2010-03-15

摘要/Abstract

摘要： 在解释变量内生条件下，Choi,Saikkonen(2004)使用动态最小二乘法估计协整平滑转移回归模型，并基于动态最小二乘的估计结果构造统计量检验协整向量的非线性。本文系统解析了的构造并指出其不足，针对这一不足，本文将动态最小二乘法扩展为完全修正的最小二乘法，并进而基于完全修正的最小二乘法估计结果构造统计量检验协整向量的非线性。本文的仿真试验表明，在有限样本下，与的检验势没有显著差异，但的水平扭曲小于。

Abstract: Under the condition that the regressors are endogenetic, Choi,Saikkonen (2004) estimate cointegrating smooth transition regressions model with DOLS, and develop statistic to test nonlinearity of cointegrating vectors. This paper elaborates how to structure and points out its shortcoming. In view of this shortcoming, the paper extend Choi, Saikkonen’s(2004)DOLS with FMOLS, and proposes statistic to test nonlinearity of cointegrating vectors in cointegrating smooth transition regressions model. The Monte Carlo simulation results show that: under finite sample, and has almost same power, but the size distortion of is greater than that of .

Key words: Cointegrating Smooth Transition Regressions Model, Dynamic OLS, Fully Modify OLS, Nonlinearity

欧阳志刚. 协整平滑转移回归中的线性检验 ——基于完全修正最小二乘法的扩展 [J]. 统计研究, 2010, 27(3): 70-75.

Ouyang Zhigang. Testing Linearity in Cointegrating Smooth Transition Regressions ——Based on Fully Modified OLS [J]. Statistical Research, 2010, 27(3): 70-75.