基于无穷状态转换模型的中国股票市场收益率分析

中国管理科学 ›› 2014, Vol. 22 ›› Issue (1): 10-19.

基于无穷状态转换模型的中国股票市场收益率分析

蔡伟宏^1,2, 唐齐鸣¹

1. 华中科技大学经济学院, 湖北武汉 430074;
2. 广东外语外贸大学国际经济贸易学院, 广东广州 510006

收稿日期:2012-02-24 修回日期:2012-07-20 出版日期:2014-01-20 发布日期:2014-01-20

Infinite Markov-switching VAR Model and Application to Analysis of China Stock Market Return

CAI Wei-hong^1,2, TANG Qi-ming¹

1. School of Economics, Huazhong University of Science and Technology, Wuhan 430074, China;
2. School of International Trade and Economics, Guangdong University of Foreign studies, Guangzhou 510006, China

Received:2012-02-24 Revised:2012-07-20 Online:2014-01-20 Published:2014-01-20

摘要/Abstract

摘要： 本文建立一个状态数目由数据决定的马尔可夫转换向量自回归模型，用贝叶斯方法推断模型参数，并利用基于Gibbs分块采样的MCMC方法做逼近。然后本文用此模型和估计方法分析上海A股市场周收益率，结果发现，我国股票市场最可能存在5个不同的状态，状态间的区分首以波动性大小不同为标准，股市除了在初期波动性极小外，从1992年4月开始可以分为两个阶段，在各阶段股市均在三个状态之间转换。

关键词: 无穷状态转换, 向量自回归, 贝叶斯推断, 分块采样, A股市场收益率

Abstract: A new Markov-switching VAR model is developed in which the number of the regimes is driven by data. The model is inferenced with Bayesian methods, and estimated with block sampling based MCMC method. With the studying of weekly return data of Shanghai A share market, five contrasted regimes are identified using the proposed model which are differenciated by return volatility. Before April 1992, market volatility is extremely low and since then there are two periods and three regimes are switching in each period.

Key words: infinite Markov-switching, VAR, Bayesian inference, block sampling, A share market return

中图分类号:

F224.0

蔡伟宏, 唐齐鸣. 基于无穷状态转换模型的中国股票市场收益率分析[J]. 中国管理科学, 2014, 22(1): 10-19.

CAI Wei-hong, TANG Qi-ming. Infinite Markov-switching VAR Model and Application to Analysis of China Stock Market Return[J]. Chinese Journal of Management Science, 2014, 22(1): 10-19.

参考文献

[1] Hamilton J. A new approach to the economic analysis of nonstationary time series and the business cycle[J]. Econometrica, 1989, (2):357-384.
[2] Hamilton J. Time series analysis[M]. New Jersey: Princeton University press, 1994.
[3] Krolzig M. Statistical analysis of cointegrated VAR processes with markovian regime shifts[R].Working Papers, Humboldt University, 1996.
[4] Cai Jun. A Markov model of switching-regime ARCH[J]. Journal of Business and Economic Statistics, 1994, 12(3):309-316.
[5] Garcia R, Perron P. An analysis of the real interest rate under regime shifts[J]. The Review of Economics and Statistics, 1996, 78(1):111-125.
[6] Mills T C, Wang Ping. Regime shifts in European real interest rates[J].Review of world economics, 2003, 139(1)66-81.
[7] 朱钧钧, 谢识予.上证综指马尔可夫转换模型的MCMC估计和分析[J].系统工程, 2010, 28(4):9-14.
[8] 朱钧钧, 谢识予.中国股市波动率的双重不对称性及其解释——基于MS_TGARCH模型的MCMC估计和分析[J].金融研究, 2011, (3): 134-148.
[9] 严太华, 陈明玉.基于马尔科夫切换模型的上证指数周收益率时间序列分析[J].中国管理科学, 2009, 17(6):33-38.
[10] Kim C J, Nelson C R. State-space models with regime-switching: Classical and gibbs-sampling approaches with applications[M].Cambridge, Mass: The MIT Press, 1999.
[11] Psaradakis Z, Sola M. Finite sample properties of the maximum likelihood estimator in autoregressive models with Markov switching[J].Journal of Econometrics, 1998, 86:369-386.
[12] Beal M J, Ghahramani Z, Rasmussen C E. The infinite hidden markov model[M].//Dietterich T G, Ghahrama Z ni. Advances in neural information processing systems. Cambridge, MA:MIT Press, 2002.
[13] Teh Y W, Jordan M I, Beal M J, et al. Hierarchical dirichlet processes[J]. Journal of the American Statistical Association, 2006, 101:1566-1581.
[14] Fox E B, Sudderth E B, Jordan M I, et al. Nonparametric bayesian learning of switching linear dynamical systems[M].//Koller D. Advances in neural information processing systems 21, Cambridge, MA:MIT Press, 2009.
[15] Carvalho C, Lopes H. Simulation based sequential analysis of Markov switching stochastic volatility models[J].Computational Statistics and Data Analysis, 2006, 51(9):4526-4542.
[16] 张兵.基于状态转换方法的中国股市波动研究[J].金融研究, 2005, (3):100-108.
[17] 高金余, 陈翔.马尔可夫切换模型及其在中国股市中的应用[J].中国管理科学, 2007, 15(6):20-25.
[18] Geweke J, Amisano G. Comparing and evaluating Bayesian predictive distributions of asset returns[J]. International Journal of Forecasting, 2010, 26(2):216-230.
[19] Kass R E, Raftery A E. Bayes factors[J]. Journal of the American Statistical Association, 1995, 90(430):773-795.
[20] Dawid A P. Statistical theory: The prequential approach[J]. Journal of the Royal Statistical Society, 1984, Ser. A, 147:278-292.
[21] Brock W A, Dechert D, Scheinkman H, et al. A test for independence based on correlation dimension[J]. Econometric Reviews, 1996, 15(3):187-235.
[22] Luginbuhl R, Vos A. Bayesian analysis of an unobserved component time series model of GDP with Markov switching and time varying growths[J]. Journal of Business and Economic Statistics, 1999, 17(4):456-465.
[23] Albert J H, Chib S. Bayes inference via gibbs sampling of autoregressive time series subject to markov mean and variance shifts[J]. Journal of Business and Economic Statistics, 1993, 11(1):1-15.
[24] Perlin M.MS_regress-the MATLAB Package for Markov regime switching models[EB/OL] . 2010-11-26.http://ssrn.com/abstract=1714016.