面板数据马尔可夫体制转换回归模型估计的EM算法及其应用  

EM Algorithm of Estimating Panel Markov Regime Switching Regressive Models and Its Application

作　　者：白仲林[1] 赵亮[1] 

机构地区：[1]天津财经大学统计系

出　　处：《数量经济技术经济研究》2012年第5期121-133,共13页Journal of Quantitative & Technological Economics

基　　金：国家自然科学基金项目"基于Bayesian方法的面板单位根检验和协整检验方法研究"(70771072);国家教育部人文社科项目"伪面板数据建模方法及其应用研究"的资助

摘　　要：本文提出了具有个体固定效应的面板数据马尔可夫体制转换回归模型及EM算法,并利用蒙特卡罗模拟方法讨论了EM算法的准确性。模拟分析发现,随着样本容量的增加,面板数据马尔可夫体制转换回归模型EM算法的估计准确性提高,增加面板数据的时期数更有助于改善估计效果。本文对10个OECD国家的经济周期性进行实证分析表明,经济在"衰退"和"扩张"两体制之间的转换存在显著的非对称性,并且世界经济正逐步强势复苏。This paper proposes a panel Markov regime switching regressive Models with individual fixed effects, PMRS models, and EM algorithm of their es- timation. We apply the Monte Carlo simulations to discuss the accuracy of the EM algorithm, The results show that, EM algorithm for PMRS models improves the accuracy of the estimates with the sample size increases, particularly to increase the number of panel period improve helpfully the effect of parameter estimations. In addition, the paper studies empirically duration of business cycles in 10 OECD countries, the results show that there is a significant asymmetry for switches be- tween the two regimes, ＂recession＂ and ＂expansion＂; moreover, the world econo- my go really out of 2008＇s financial crisis and recovery gradually.

关 键 词：面板数据 马尔可夫模型 EM算法 蒙特卡罗模拟 

分 类 号：F224.0[经济管理—国民经济]