多元时间序列GARCH型模型的诊断检验

数学物理学报 ›› 2010, Vol. 30 ›› Issue (3): 630-638.

多元时间序列GARCH型模型的诊断检验

¹吴鑑洪|²朱力行

1.浙江工商大学统计与数学学院杭州 310018|
2.香港浸会大学数学系香港

收稿日期:2008-12-11 修回日期:2009-12-23 出版日期:2010-05-25 发布日期:2010-05-25
基金资助:
香港研究资助局项目、中国教育部人文社科项目(07JJD790154)、浙江省自然科学基金(Y6090172)以及浙江工商大学青年人才基金资助

Diagnostic Checking for Multivariate GARCH-type Models in Time Series

¹ WU Jian-Hong, ²ZHU Li-XIng

1.School of Statistics and Mathematics, Zhejiang Gongshang University, Hangzhou 310018;
2.Department of Mathematics, |Hong Kong Baptist University, Hong Kong

Received:2008-12-11 Revised:2009-12-23 Online:2010-05-25 Published:2010-05-25
Supported by:
香港研究资助局项目、中国教育部人文社科项目(07JJD790154)、浙江省自然科学基金(Y6090172)以及浙江工商大学青年人才基金资助

摘要/Abstract

摘要：

多元时间序列GARCH型模型已被证实在理论和实际中具有重要作用. 该文对这一类模型的拟合优度提出了一组得分型检验统计量. 这些检验在零假设模型下渐近服从卡方分布, 计算简单, 临界值容易得到. 检验对备择模型比较敏感, 能侦察到以$1/\sqrt n$的速度收敛到零假设的备择模型. 对于可能的多个备择, 构造了渐近分布自由的Maximin检验; 而对于饱和备择情形, 基于得分型检验的思想提出了一个构造Omnibus检验的可能性. 值得指出的是构造的这组检验能检测到零假设模型的条件协差阵的每一部分可能的偏离, 从而当模型被错误指定时, 该检验能提供相关信息进行模型修正. 模拟结果表明该文的检验表现理想.

关键词: GARCH -型模型, Maximin 检验, 模型诊断检验, 得分型检验

Abstract:

Multiple time series with time-varying conditional variance are verified to be useful in both theory and real applications. In this article, a battery of score-type diagnostic tests are suggested for checking the adequacy of multivariate GARCH-type models fitting multiple time series. The
resulting tests are asymptotically chi-squared under the null hypothesis, and critical values can be obtained easily. Moreover, they are sensitive to alternatives and can detect the alternatives approaching the null at the rate arbitrarily close to 1/√n, the fastest possible rate in goodness-of-fit test. For a large number of alternatives, an asymptotically distribution-free maximin test is constructed. A possibility to construct score-based omnibus tests is discussed when the alternative is saturated. Furthermore, it is worthwhile to show that the battery of tests can detect separate aspects of possible departures from the null and then provide us more specifical information about the likely source of misspecification. Simulation results indicate that the tests perform well.

Key words: GARCH-type models, Maximin test, Model diagnostic checking, Score type test

中图分类号:

62F05

吴鑑洪, 朱力行. 多元时间序列GARCH型模型的诊断检验[J]. 数学物理学报, 2010, 30(3): 630-638.

WU Jian-Hong, ZHU Li-XIng. Diagnostic Checking for Multivariate GARCH-type Models in Time Series[J]. Acta mathematica scientia,Series A, 2010, 30(3): 630-638.

参考文献

[1] Bauwens L, Laurent S, Rombouts J. Multivariate GARCH models: A survey. J Appl Econ, 2006, 21: 79--109

[2] Billingsley P. The Lindeberg-Levy theorem for martingales. Proc Amer Math Soc, 1961, 12: 788--792

[3] Bollerslev T, Engle R F, Wooldridge J M. A capital asset pricing model with time varying covariance. J Polit Econ, 1988, 96: 116--131

[4] Bollerslev T, Wooldridge J M. Maximum likelihood estimation and inference in dynamic models with time varying covariance. Economet Rev, 1992, 11: 143--172

[5] Engle R F. Autoregressive conditional heteroscedasticity with estimates of the variance of UK inflation. Econometrica, 1982, 50: 987--1008

[6] Engle R F, Grange C W J, Kraft D. Combining competing forecasts of inflation using a bivariate ARCH model. J Econ Dyna Cont, 1984, 8: 151--165

[7] Jensen S T, Rahbek A. Asymptotic inference for nonstationary GARCH, Econometric Theory, 2004, 20(6): 1203--1226

[8] Kraft D F, Engle R F. Autoregressive Conditional Heteroscedasticity in Multiple Time Series. Unpublished Manuscript. San Diego, CA: Department of Economics, University of California, 1983

[9] Lee S, Hansen B. Asymptotic theory for the ARCH(1, 1) quasi-maximum likelihood estimator. Econometric Theory, 1994, 10: 29--52

[10] Li W K. Diagnostic Checks in Time Series. Netherlands: Chapman and Hall, 2004

[11] Li W K, Mak T K. On the squared residual autocorrelations in non-linear time series with conditional heteroscedasticity. J Time Ser Anal, 1994, 15: 627--636

[12] Ling S, Li W K. Diagnostic checking of nonlinear multivariate ARCH errors. J Time Ser Anal, 1997, 18: 447--464

[13] Ling S, McAleer M. Asymptotic theory for a vector ARMA-GARCH model. Econometric Theory, 2003, 19: 280--310

[14] Lundbergh S, Teräsvirta T. Evaluating GARCH models. J Econom, 2002, 110: 417--435

[15] Strasser H. Mathematical Theory of Statistics. Berlin: De Gruyter, 1985

[16] Stute W, Zhu L X. Nonparametric checks for single-index models. Ann Statist, 2005, 33(3): 1048--1083

[17] Tse Y K. Residual-based diagnostic for conditional heteroscedasticity model. Econom J, 2002, 5: 358--373

[18] Wang M, Yao Q. Modelling Multivariate Volatilities: an ad hoc Approach. In Contemporary Multivariate Analysis and Experimental Designs -- In Celebration of Professor Kai-Tai Fang's 65th Brithday. Singapore: World Scientific, 2005: 87--97

[19] Wu J H, Zhu L X. A score type test for general autoregressive models in time series. Acta Math Appl Sin, 2007a, 23(3): 439--450

[20] Wu J H, Zhu L X. Empirical likelihood ratio tests for multivariate regression models. Front Math China, 2007b, 2(1): 149--168

[21] Wu J H, Zhu L X. Testing the adequacy of GARCH-type models in time series. Acta Mathematica Scientia Ser B, 2009a, 29B(2): 327--340

[22] Wu J H, Zhu L X. Goodness-of-Fit tests for vector autoregressive models in time series. Sci China Ser A, 2009b, 52, DOI: 10.1007/sll 425-009-0071-1

[23] Zhu L X, Cui H J. Testing the adequacy of a general linear error-in-variables model. Statist Sinica, 2005, 15: 1049--1068