摘要
世界金融衍生品市场迅猛发展,它在为参与者提供避险工具的同时,也成为金融市场发生剧烈波动的根源,增大了风险管理的复杂性。对金融风险进行有效的管理成为国内外金融实务界、理论界和监管机构共同关注的焦点。风险测量是风险管理中首要而核心的部分,研究风险测量对于风险管理理论以及我国风险管理的发展具有十分重要的理论意义和现实指导意义。
VaR和CVaR方法在风险测量、监管等领域获得广泛应用,成为金融市场风险测量的主流方法,基于GARCH类模型的参数估计方法是其中很重要的方法之一。股票价格的波动呈现出非对称性、非均衡的关系,收益的均值和方差是自相关的、不稳定的。其中非对称性与投资者的心理因素密切相关,然而,目前的这些非对称GARCH模型很少考虑投资者的心理因素对股价运行规律的影响。
本论文分析已有的GARCH类模型,结合行为金融学中的前景理论,构建基于价值函数的VF-GARCH模型,并给出参数估计的极大似然函数估计方法。结合风险管理测量方法中的VAR和CVAR理论,运用VF-GARCH模型改进VAR和CVAR测量方法。选择沪深300指数的对数收益率作为样本,分别采用VF-GARCH模型和GARCH模型对沪深300指数的对数收益率进行建模,比较得出VF-GARCH模型具有优越性。利用VF-GARCH模型改进的VAR和CVAR参数方法测量沪深300指数对数收益率的VAR和CVAR值,采用Kupiec的失败频率检验法对测量结果进行有效性检验,实证结果对于沪深300指数期货的风险管理具有指导意义。
With the rapid development of the world's financial derivatives market, it provide participants with hedging tool, has become the root causes of volatile financial markets, increased the complexity of risk management. The effective management of financial risks has become focus attention of domestic and foreign financial practitioners, theoretical circles and regulatory agencies. Risk measurement is the core part of risk management. The study of risk measure has great theoretical significance and practical significance for risk management theory and the development of China's risk management.
VaR and CVaR method is widely applied in risk measurement, monitoring and other fields .It becomes the mainstream of financial market risk measurement methods and parameter estimation method based on the GARCH class model is one of the very important way. The volatility of stock prices showing a non-symmetrical, non-equilibrium and relationship between the mean and variance of income is self-related, and unstable. In which the non-symmetry are closely related to investors psychological factors, however, these non-symmetric GARCH model were with little regard for investors psychological factors on the price impact at present.
This paper analyzes the existing GARCH class model, combined with the prospect theory of behavioral finance theory, build VF-GARCH model based on value function, and gives the maximum likelihood function parameter estimation method. Combined with measurement methods of risk management in the VAR and CVAR theory, it improved the VAR and CVAR measurement methods which based on VF-GARCH model. Select the number of the Shanghai and Shenzhen 300 Index log-return rate as a sample, using number of the Shanghai and Shenzhen 300 Index rate of return on the VF-GARCH and GARCH modeling, and comparison of derived VF-GARCH model has the advantage. Use VAR and CVAR parameters methods which improved based on VF-GARCH model to measure the rate of log-return of the Shanghai and Shenzhen 300 index VAR and CVAR values, measure the effectiveness of the results with the Kupiec’s failure frequency test method, the empirical results of the Shanghai and Shenzhen 300 index futures have guiding significance of risk management.
引文
[1] Engel. Auto regressive conditional hetero skedasticity with estimates of U. K inflation [J]. Econometrics, 1982, 50: 987- 1008.
[2] Bollerslew Tim. Generalized Auto regressive Conditional hetero scedasticity [J]. Journal of Econometrics, 1986,31-32.
[3] Hsieh D. A. (1989): Testing for nonlinear dependence in daily foreign exchange rates[J]. Journalof Business 62: S.339-368.
[4] French, Kenneth R, Cz William Schwert and Robert F. Stambaugh, 1987: Expected Stock Returns and Volatility[J],Journal of Financial Economics, September, pp. 3-30.
[5] Engle, R.F., Lilien, D.M. and Robbins, R.P. (1987) Estimating Time Varying Risk Premia in the Term Structure: the ARCH-M Model". Econometrical [J], 55,391-408.
[6] Nelson, D., 1990: ARCH models as diffusion approximations[J],Journal of Econometrics,45, pp,7-39.
[7] Nelson, D.B. (1991)"Conditional Heteroskedasticity in Asset Returns”.Econometrical [J],59,347-70.
[8] Glosten, L.R., Jagannathan,R. and R and Runkle D. (1993) "Relationship Between the Expected Value and the Volatility of the Nominal Excess Return on Stocks". Journal of Finance[J],48,1779-801.
[9]万建强,文洲. ARIMA模型与ARCH模型在香港股指预测方面的应用比较[J] ,数理统计与管理, 2001,5,1-4.
[10]许爱霞.GARCH模型对沪市行业指数的实证研究[J].市场论坛.2006,3.108-109.
[11]金丹.沪深300指数的ARCH效应[J].经济学研究.2008,3.47-51.
[12]吴子荣.沪深300指数的时间序列分析[J].西南金融.2007,9.47-48.
[13]严定琪,李育锋.基于GARCH族模型的沪深300指数波动率预测[J].兰州交通大学学报2008,2.92-95.
[14]张虹.基于NN-GARCH模型的中国沪深300指数实证研究[D].天津财经大学硕士学位论文.5-43.
[15]章蓉.VFB-GARCH模型及其研究[D].哈尔滨工业大学硕士学位论文.15-38.
[16]何帮强,惠军.基于ARMA-EGARCH-M模型的沪深股市波动性分析[J].合肥工业大学学报.2007,7.865-868.
[17] Rockafeller T, Uryasev S.Optimization of conditional value-at-risk, Journal of Risk 2000,2(3): 21-24.
[18] Danielsson J.,Jorgensen B.N.,and de Vries C.G. Risk Measures and Investment Choice.canbe download from: www.RiskResearch.org. 2001b.
[19] Antonelli,S.,Gabriella,M. Optimization of Monte Carlo Procedures for Valueat Risk Estimates.Economic Notes Banca Monte dei Paschi di Siena SpA.2002,31(1). 59-78.
[20] Andrey Rogachev. Dynamic Value-at Risk. can be downloaded from: www.gloriamundi.org. 2002.
[21] Cabedo Sember J.David,Moya Clemente Ismael. Estimating Oil Price "Valueat Risk" Using the Historical Simulation Approach. can be downloaded from: www. gloriamundi.org. 2003 ( 5 ) .
[22]王树娟,黄渝祥.基于GARCH-CVaR模型的我国股票市场风险分析[J].同济大学学报,2005,2.260-263.
[23]邱沛光.上证指数风险特征实证分析[J].陕西农业科学,2005,1.71-74.
[24]杨树成.深证100指数的风险值估计方法[J].重庆文理学院学报,2008,2.15-18.
[25]刘静.我国股价指数风险价值实证分析[J],.经济问题探索2002,3.70-74.
[26]杜红军.VaR和CVaR风险值的估计和计算[D].华中科技大学硕士学位论文,2006.21-45.
[27]高可佑,王潇怡,黄勇兵.沪深300指数的VaR风险测量—基于历史模拟法和蒙特卡罗模拟法[J].市场周刊.2008,3.90-91.
[28]周孝华,唐秋燕.沪深300指数极值VaR的分析与计算[J],统计与决策.2008,10.96-98.
[29]周小敏.基于GARCH模型的CVaR金融风险测度研究[D],湖南大学硕士学位论文,2007.12-28.
[30]徐静,基于GJR-GARCH模型和极值理论的VaR评估[D].东北财经大学硕士学位论文.2006,27-40.
[31]李心丹.行为金融学—理论及中国的证据.2003:145-167.
[32] S.Hersh,M.Statman.Explaining Investor Preference for Cash Dividends.Journal of Financial Economics.1984,13(2):253-282.
[33] S.Hersh,Beyond Greed and Fear:Understanding Behavioral Finance and thePsychology of Investing.Boston.MA:Harvard Business School Press,2000:67-98.
[34] Daniel,Hirschleifer and Subrahmanyam.Investor Psychology and SecurityUnder-and Overreactions.Journal of Finance.1999:7-14.
[35] H.Stein.A Unified Theory of Underreaction,Momentum Trading and Overreaction in Asset Markets.Journal of Finance.1999:42-67.
[36] R.Bloomfield,R.Libby and M.W.Nelson.Underreactions,Overreactions and Moderated Confidence.Journal of Financial Markets.2000,(3):113-137.
[37] J.Kang,M.H.Liu and S.X.Ni.Contrarian and Momentum Strategies in the China Stock Market:1993-2000.Pacific-Basin Finance Journal.2002,(10):243-265.
[38] A.M.Poteshman.Underreaction,Overreaction,and Increasiing Misreaction toInformation in the Options Market.The Journal of Finance.2001,(3):26-45.
[39] S.J.Larson,J.Madura.Overreaction and Under reaction In the Foreign Exchange Market.Global Finance Journal.2001,(12):153-177.
[40] B.Ahmet,C.Nusret.Do markets overreact International evidence. Journal of Banking and Finance.1999,(23):1121-1144.
[41] R.P.Chang,D.W.McLeavey and S.G..Rhee.Short-term abnormal returns of the contrarian strategy in the Japanese stock market.Journal of Business Finance and Accounting.1995,(22):1035-1048.
[42] A.Hameed,S.Ting. Trading volume and short-horizon contrarian profits:Evidence from Malaysian stock market. Pacific-Basin Finance Journal.2000,(8):67-84.
[43] M.Theobald,P.Yallup.Determining Securiy Speed of Adjustment Coefficients. Journal of Financial Markets.2002:251-252.
[44]韩海平,陶燕红,方兆本.行为金融学综述[J].经济研究.2002,(1):142-143.
[45]黄树青.行为金融学与传统金融学争论[J].经济学动态.2002,(1):91-125.
[46]翁学东.西方行为金融学理论的进展[J].中央财经大学学报,2003.(1):35-59.
[47]张俊喜,张华.解析我国封闭式基金折价之谜[J].金融研究.2002,(12):58-66.
[48]沈洪溥,周立群.行为金融学对封闭式基金价格波动的解释[J].金融教学与研究.2003,(5):98-102.
[49]饶育蕾,杨琦.我国封闭式基金折价交易的行为金融学实证分析[J].中南人学报2003,(3):90-92.
[50]吴忠群.ST类股票价格对公司财务状况的反应[J].经济与管理研究.2002,(6):7-9.
[51]王永宏,赵学军.中国股市“惯性策略”和“反转策略”的实证分析[J].经济研究.2001,(6):33-41.
[52]吴世农,吴超鹏.我国股票市场“价格惯性策略”和“赢余惯性策略”的实症研究[J].经济科学.2003,(4):68-69.
[53]何诚颖.中国股市“板块现象”分析[J].经济研究.2001,(12):82-91.
[54]赵学军,王永宏.中国股市“处置效应”的实证分析[J].金融研究.2001,(7):33-34.