变结构面板协整检验的响应面函数研究

The Response Surface Function of Panel Cointgeration Test with Breaks

Panel-ρ检验是常用的异质面板数据协整检验统方法,本文通过Monte Carlo模拟分析Panel-ρ协整检验方法的小样本性质,以及该方法对于变结构面板协整检验的检验功效.结果表明,在小样本情况下Panel-ρ统计量在原假设下的渐进分布会不同于该统计量的理论渐进分布,而且面板数据中存在的结构变化也会对渐进分布产生影响,从而降低Panel-ρ检验的检验功效.为了得到更加符合实际样本情况的统计量临界值,通过Monte Carlo模拟方法得到Panel-ρ协整检验方法的响应面函数,建立了统计量的临界值与面板数据的样本容量、结构变化类型的直接函数关系.Monte Carlo模拟检验表明,响应面函数法确实能够改善Panel-ρ协整检验的检验功效,在小样本容量和具有结构变化的情况下保证了面板协整检验的有效性.

Abstract： Panel-ρ statistic is a regular method for uses Monte Carlo simulation to analyze its test power heterogeneous panel cointegration test. This paper in small samples and in structural changes condition. The results show that the asymptotic distribution of Panel-ρstatistic under the null hypothesis will be influenced by the small samples case and the structural changes. In order to get an appropriate critical values, this paper applies the response surface function to fit the critical values which obtained by Monte Carlo simulation. The Monte Carlo simulations prove that the response surface functions can obviously improve test power of the Panel-ρ method, make the Panel-ρ method more effective with small samples and the structural changes.