本文基于Copula方法对由高频分笔数据得到的交易量持续期进行了研究。应用多元藤Copula方法对连续几个交易量持续期之间的自相依结构进行估计,在此基础上提出了一种新的条件密度函数估计方法,进而给出了交易量持续期的预测。对中国石化高频分笔数据进行实证分析的结果表明,本文模型对持续期的预测能力要明显优于EACD模型,在密度函数预测检验方面,本文模型也有更好的表现。
In this paper, the trading volume duration sequence derived from high-frequency tick-by-tick data is analyzed by Copula method. The auto-dependence structure of several consecutive trading volume durations is estimated by multivariate vine Copula, then, a new estimating method about conditional density function forecasting is also proposed. Moreover, a new forecasting method of the volume duration is put forward. Empirical results of Sinopec show that the predictive ability of our model is much better than that of EACD, which can also be demonstrated from the density forecasting test.
[1] Diamond D W, Verrecchia R E. Constraints on short-selling and asset price adjustments to private information[J]. Journal of Financial Economics, 1987, 82(2):33-53.
[2] Engle R F, Russell J R. Autoregressive conditional duration:A new approach for irregularly spaced transaction data[J]. Econometrics, 1998, 66(5):1127-1162.
[3] Lunde A. A generalized gamma autoregressive conditional duration model[R]. Working Paper, Department of Economics, University of Aarhus, 1998.
[4] Grammig J, Maurer K O. Non-monotonic hazard functions and the autoregressive conditional duration model[J]. The Econometrics Journal, 2000, 3(1):16-38.
[5] Bauwens L, Giot P. The logarithmic ACD model:An application to the bid-ask quote process of three NYSE stocks[J]. Annalesd Economie et de Statistique, 2000, 60:117-149.
[6] Savu C, Ng W L. The SCoD Model:Analyzing durations with a semi-parametric copula approach[J]. International Review of Finance, 2005, 5(1-2):55-74.
[7] Ning C. Dependence structure between the equity market and the foreign exchange market-A copula approach[J]. Journal of International Money and Finance, 2010, 29(5):743-759.
[8] Bedford T, Cooke R M. Probability density decomposition for conditionally dependent random variables modeled by vine[J]. Annals of Mathematics and Artificial Intelligence, 2001, 32(1):245-268.
[9] Aas K, Czado C, Frigessi A, et al. Pair-copula constructions of multiple dependence[J]. Insurance:Mathematics and Economics, 2009,44(2),182-198.
[10] Heinen A, Valdesego A. Asymmetric CAPM dependence for large dimension:The canonical vine autoregressive copula model.Working Paper, SSRN,2009.
[11] Joe H, Li Haijun. Tail dependence functions and vine copulas[J]. Journal of Multivariate Analysis, 2010, 101(1):252-270.
[12] Righi M B, Ceretta P S.Analyzing the dependence structure of various sectors in the Brazilian market:A pair copula construction approach[J]. Economic Modelling, 2013, 35:199-206.
[13] Bauwens L, Giot P, Grammig J, et al. A comparison of financial duration models via density forecasts[J]. International Journal of Forecasting, 2004, 20(4):589-609.
[14] Skalr A. Fonctions de repartition and dimensions et leurs marges[J].Publications de l'lnstitnt Statistique de l' Universite de Paris, 1959,(8):229-231.
[15] 李广川,刘善存,邱菀华.交易量持续期的模型选择:密度预测方法[J].中国管理科学,2008,16(1):131-141.
[16] Diebold F X, Gunther T A, Tay A S. Evaluating density forecasts with applications to financial risk management[J]. International Economic Review, 1998, 39(4):863-883.
[17] 叶五一,缪柏其,吴遵.基于分位点自回归模型的动态持续期风险估计[J],数理统计与管理,2010,29(3):500-517.
