-118 系统工程学报 第21卷 Kurt=19.9876 kew=2.5438 (20) 4结论 从上面的结果我们可以看出,模拟收益序列的蜂度 和偏度远远大于正态分布的峰度和偏度,从而也就 本文中,基于Euler方法获得参数后验分布的 证明了双指数跳跃扩散模型能够很好地同时描述 离散密度,用McMC方法估计了双指数跳跃扩散 资产收益分布的尖峰厚尾和有偏性的特征.另外, 模型及模型中的隐含跳跃变量,发现McMC方法 在用MMC方法对参数进行估计的同时,还估计了 可以很好地估计连续时间模型,特别是含有隐含 跳跃发生的时间和跳跃的规模,而Kou在估计双指 变量的连续时间模型.通过该模型所模拟的股指 数跳跃扩散模型的时候仅仅估计了模型的参数.所 数据和实际数据的分布基本一直,从而说明了 以,模拟结果和估计结果也说明了MMC方法适合 DED扩散可以很好的描述资产收益的分布特征, 估计含有隐含变量的连续时间模型, 参考文献: [1]Merton R C.Theory of rational option opricing[J].Bell Joumal of Economics,1973,4(1):141-183 [2]Black F,Scholes M.The pricing of options and corporate liabilities[J].Joumal of Political Economy,1973,81(3):637-659 [3]Bamdorff N,Shleifer A,Vishny R.A model of investor sentiment[J].Joumal of Financial Economics,1998,49:307-343. [4]Eberlein E,Keller U.Hyperbolic distribution in finance[J].Bemoulli,1995,3(1):281-299. [5]Madan D B,Carr P,Chang EC.The variance gamma process and option pricing[J].European Finance Review,1995,2:79-105. [6]Adersen L,Andreasen J.Jump-diffusion process:Volatility smile fitting and numerical methods for pricing[].Review of Deriva- tives Research,2000,4:231-262. [7]Hull JC,White A.The pricing of options on assets with stochastic volatilities[J].Joural of Finance,1987,42(1):281-300. [8]Cox J C,Ross S.The Valuation of options for alternative stochastic processes[J].Joumal of Financial Economics,1976,3(3): 145-166. [9]Robert M C.Option pricing when underlying stock returns are discontinuous[J].Joumal of Financial Economics,1976,3(3): 125一144. [10]Kou SG.A jump-diffusion model for option pricing[J].Management Science,2002,48(8):1086-1101. [11]Asger P.A new approach to maximum likelihood estimation for stochastic differential equations based on discrete obeervations[]. Scandinavian Joumal of Statistics,1995,22(1):55-71. [12]Eraker B.MeMC analysis of diffusion models with application to finance[].Joumal of Business and Economic Statistics,2001,19 (2):177191. [13]Elerian O,Chib S,Shephard N.Likelihood inference for discretely observed non-linear diffusions[].Econometrica,2001,69 (4):959-993. [14]Tse Y K.Zhang X B,Yu J.Estimation of hyperbolic diffusion using the Markov chain Monte Carlo method[J].Quantitative Fi- nance,2003,3(1):1-12. [15]胡素华,张形,张世英.正态逆高斯扩散模型的McMC估计[J】.系统工程理论方法应用,2005,15(2):41一46. [16]Chibs S,Greenberg E.Understanding the metropolis-hastings algorithm[J].American Statistician,1995,49(2):327-35. [17]Gilks WR,Richardson S,Spiegelhatler D J.Introducing Markov chain Monte Carlo[A].In:Markov Chain Monte Carlo in Prac- tice[M].Gilks WR,Richardson S,Spiegelhalter D J.London:Chapman and Hall,1996.1-20. [18]Roberts GO.Markov chain concepts related to sampling algorithms[A].In:Markov Chain Monte Carlo in Practice[M].Gilks W R,Richardson S,Spiegelhalter D J.London:Chapman and Hall,1996.45-57. [19]Kim S,Shephard N,Chib S.Stochastic volatility:Likelihood inference and comparison with ARCH models[J].Review of Eco- nomic Studies,1998,65(2):361—93. [20]Meyer R,Yu J.BUGS for a Bayesian analysis of stochastic volatility models[J].Econometrics Joumal,2000,3(2):19-215. 作者简介: 胡素华(197一),男,安徽怀宁人,博士生,讲师,研究方向:连续时间金融模型研究,金融波动研究; 张世英(1936一),男,北京人,教授,博士生导师,研究方向:社会经济系统建模与控制; 张形(1968一),女,天津人,副数授,硕士生导师,研究方向:金融分析,公司理财,资产评估. 万方数据一118一 系统工程学报 第21卷 Kurt=19.987 6 Skew=2.543 8 (20) 从上面的结果我们可以看出,模拟收益序列的峰度 和偏度远远大于正态分布的峰度和偏度,从而也就 证明了双指数跳跃扩散模型能够很好地同时描述 资产收益分布的尖峰厚尾和有偏性的特征.另外, 在用McMC方法对参数进行估计的同时,还估计了 跳跃发生的时问和跳跃的规模,而Kou在估计双指 数跳跃扩散模型的时候仅仅估计了模型的参数.所 以,模拟结果和估计结果也说明了McMC方法适合 估计含有隐含变量的连续时间模型. 参考文献: 4结论 本文中,基于Euler方法获得参数后验分布的 离散密度,用McMC方法估计了双指数跳跃扩散 模型及模型中的隐含跳跃变量,发现McMC方法 可以很好地估计连续时间模型,特别是含有隐含 变量的连续时间模型.通过该模型所模拟的股指 数据和实际数据的分布基本一直,从而说明了 DEJD扩散可以很好的描述资产收益的分布特征. [1 j Menon R C.Theory of rational option opricing[J].Bell Journal of Economics,1973,4(1):141—183. [2]Black F,Scholes M.The pricing of options and corporate liabilities[J].Journal of Political Economy,1973,81(3):637-_659. [3]Barndorff N,Shleifer A,Vishny R.A model of investor sentiment[J].Journal of Financial Economics,1998,49:307—343. [4]Ebedein E,Keller U.Hyperbolic distribution in finance[J].Bernoulli,1995,3(1):281—299. [5]Madan D B,Cart P,c}lang E C.The variance gmlrna process and option砸c崦[J].European Finance Review,1995,2:79—1c15. [6]Adersen L,Andreasen J.Jump-diffusion process_,Volatility smile fitting and numerical methods for pficing[Jj.Renew of Deriva— fives Research,2000,4:231—262. [7]Hull J C,碱te A.ne pricing of options on assets with stochastic volatilities[J].Journal of Finance,1987,42(1):281—300. [8]Cox J C,Ross S.‰Valuation of options for alternative stochastic processes[J].Journal of Financial Economics,1976,3(3): 145—166. [9]Robert M C.Option pricing when underlying stock returns are discontinuous[J].Journal of Financial Economics,1976,3(3): 125—144. [10]Kou S G.A jump.-diffusion model for option pricing[J].Management Science,2002,48(8):108卜1101. [11]Asger P.A new appmach to maximum likelihood estimation for stochastic differential equations based on discrete observations[JJ. Scandinavian Journal of Statistics,1995,22(1):55—71. [12]Eraker B.McMC analysis of diffusion models with application to finance[Jj.Joumal of Business and Economic Statistics,2001,19 (2):177—191. [13]Elerian O,Chib S,Shephard N.Likelihood inference for discretely observed non-linear diffusionslJj.Econometfica,2001,69 (4):959---993. [14]Tse Y K,Zhang X B,Yu J.Estimation of hyperbolic diffusion using the Markov chain Monte Carlo methodlJj.Quantitative Fi— nance,2003,3(1):1--12. [15]胡素华,张彤,张世英.正态逆高斯扩散模型的McMC估计[J].系统工程理论方法应用,2005,15(2):41—46. [16]Chibs S,Greenberg E.Understanding the metropolis-hastings algorithm[J].American Statistician,1995,49(2):327—35. [17]Gilks w R,Richardson S,Spiegelhatler D J.Introducing Markov chain Monte CarlolA].In:Markov Chain Monte Carlo in Prac— flee[M].Gilks W R,Richardson S,Spiegelhalter D J.London:Chapman and Hall,1996.1—20. [18]Roberts G O.Markov chain concepts related to sampling algofithms[A].In:Markov Chain Monte Carlo in Practice[MJ.Gilks W R,Richardson S,Spiegelhaher D J.London:Chapman and Hall,1996.45—57. [19]Kim S,Shephard N,Chib S.Stochastic volatility:Likelihood inference and comparison with ARCH modelslJJ.Review of Eco— nomic Studies,1998,65(2):361—-93. [20]Meyer R,Yu J.BUGS for a Bayesian analysis of stochastic volatility nHlelslJ].Econometrics Journal,2000,3(2):198—215. 作者简介: 胡素华(19r77一),男,安徽怀宁人,博士生,讲师,研究方向:连续时间金融模型研究,金融波动研究; 张世英(1936一),男,北京人,教授,博士生导师,研究方向:社会经济系统建模与控制; 张彤(1968一),女,天津人,副教授,硕士生导师,研究方向:金融分析,公司理财,资产评估. 万方数据