蔣致遠(yuǎn)等
摘 要 提供了一種基于自適應(yīng)拉普拉斯變換有限差分方法來(lái)解決BlackScholes 期權(quán)定價(jià)問(wèn)題.相比較于傳統(tǒng)的時(shí)間推進(jìn)法,此方法在保證較高精確度和很好的收斂性的同時(shí),還可以減少計(jì)算時(shí)間.這一精確有效的方法將通過(guò)數(shù)值實(shí)驗(yàn)來(lái)驗(yàn)證.
關(guān)鍵詞 拉普拉斯變換;有限差分;BlackScholes方程;歐式期權(quán)
中圖分類號(hào) F224.9 文獻(xiàn)標(biāo)識(shí)碼 A
Laplace Transform Based Finite Difference
Method for BlackScholes Option Pricing
JIANG Zhiyuan1, ZHANG Tiao2, GONG Shanshan2
(1. School of Business, Guilin University of Electronic Technology,Gulin, Guangxi; 2.School of Mathematic
and Computing Science, Guilin University of Electronic Technology, Guilin, Guangxi 541004 China)
Abstract This paper provids an adaptive Laplace transform finite difference method to solve the problem of BlackScholes option pricing. Comparing to the traditional time marching methods, this method not only can guarantee higher accuracy and very good convergence, but also can reduce the computation time, whose accuracy and efficiency are shown by numerical experiments.
Key words Laplace transform; finite difference; BlackScholes equation; European option
1 引 言
金融期權(quán)因?yàn)榫哂刑桌鸵?guī)避風(fēng)險(xiǎn)的功能,所以在金融市場(chǎng)占據(jù)越來(lái)越重要的位置.在國(guó)外期權(quán)市場(chǎng)飛速發(fā)展的同時(shí),國(guó)內(nèi)的期權(quán)市場(chǎng)仍處于停滯狀態(tài).同時(shí)由于國(guó)內(nèi)沒(méi)有期權(quán)交易所,導(dǎo)致部分投資者對(duì)期權(quán)鮮有了解.期權(quán)定價(jià)是期權(quán)交易的核心.在經(jīng)典的確定性BS期權(quán)定價(jià)模型中[1],為了研究的方便,人們經(jīng)常會(huì)選擇一個(gè)常數(shù)隨機(jī)波動(dòng)率.然而,這在經(jīng)典的定價(jià)模型中是有缺陷的,比如波動(dòng)率微笑就是眾所周知的一個(gè)偏差.相對(duì)于給定一個(gè)常數(shù)波動(dòng)率的經(jīng)典BS模型來(lái)說(shuō),在實(shí)際的案例中,更多的是需要來(lái)求解帶有未知波動(dòng)率的模型.于是,關(guān)于求解此類問(wèn)題的方法逐漸的受到越來(lái)越多的學(xué)者關(guān)注并研究.很多學(xué)者都采用對(duì)時(shí)間進(jìn)行積分的時(shí)間推進(jìn)法來(lái)進(jìn)行研究期權(quán)的定價(jià)問(wèn)題.其中Chawla[2]對(duì)BS方程運(yùn)用廣義梯形公式法對(duì)期權(quán)定價(jià);Vázquez[3]給出了一種逆風(fēng)數(shù)值方法的期權(quán)定價(jià)模型,可以同時(shí)對(duì)美式期權(quán)和歐式期權(quán)進(jìn)行定價(jià).國(guó)內(nèi)學(xué)者張鐵[4,5]給出了變網(wǎng)格差分方法和有限差分方法求解期權(quán)定價(jià)問(wèn)題;蹇明[6]則利用了五點(diǎn)式混合差分方法研究歐式看漲期權(quán)定價(jià)問(wèn)題.時(shí)間推進(jìn)法采用了分割足夠小的步長(zhǎng)來(lái)保持其穩(wěn)定性,但是在解決問(wèn)題的同時(shí),增加了計(jì)算量.基于這一問(wèn)題,本文運(yùn)用了自適應(yīng)拉普拉斯變換有限差分方法,這種方法是通過(guò)對(duì)時(shí)域采用拉普拉斯變換,先剔除掉暫時(shí)的衍生品,再在資產(chǎn)的價(jià)格域上對(duì)變換后得到的方程采用有限差分方法,最后利用修正后的快速拉普拉斯逆變換,以期得到依賴于時(shí)間的期權(quán)價(jià)值.
2 知識(shí)回顧
圖2給出了拉普拉斯變換有限差分方法的收斂性分析,取X=50,K=70,r=0.05,σ=0.4,T=1,L=100時(shí),通過(guò)MATLAB可以觀測(cè)到其收斂性的直觀結(jié)果,且從運(yùn)行結(jié)果上來(lái)看,其波動(dòng)性較小、比較穩(wěn)定,收斂性很好。如圖2所示:
節(jié)點(diǎn)數(shù)/個(gè)
圖2 拉普拉斯變換有限差分方法的收斂性分析
6 總 結(jié)
本文首先給出了拉普拉斯變換有限差分方法的理論推導(dǎo),并給出了相應(yīng)的算法總結(jié).最后通過(guò)數(shù)值實(shí)驗(yàn),可以觀測(cè)到,拉普拉斯有限差分法相比較于二叉樹(shù)方法和隱式差分方法,在保證其較好的收斂性的同時(shí),還有著較高的精確度.在收斂性方面其優(yōu)于隱式差分方法,但比二叉樹(shù)方法的收斂性要差一些;從計(jì)算速度上來(lái)進(jìn)行比較,其計(jì)算速度比二叉樹(shù)方法和隱式差分方法的速度要更快一些.同時(shí)需要指出該方法的不足之處,即在節(jié)點(diǎn)數(shù)的選取上,并不是節(jié)點(diǎn)數(shù)L選取的越大,其精確度就越高.所以,接下來(lái)的工作主要是研究如何選取恰當(dāng)?shù)墓?jié)點(diǎn)數(shù)以進(jìn)一步提高其精確度以及良好的收斂性.
參考文獻(xiàn)
[1] F BLACK, M SCHOLES. The pricing of options and corporate liabilites[J]. Journal of Political Economy.1973,81(3): 637-654.
[2] M M CHAWLA, M A AZANAIDI, D J EVANS. Generalized trapezoidal formulas for the BlackScholes equation of option pricing [J].Compute Math.2003,12(1):1521-1526.
[3] C VAZQUEZ. An upwind numerical approach for an American and European option pricing model[J].Applied Mathematics and Computation.1998,97(9):273-286.
[4] 張鐵,祝丹梅.美式期權(quán)定價(jià)問(wèn)題的變網(wǎng)格差分方法[J].計(jì)算數(shù)學(xué),2008,30(4):379-387.
[5] 張鐵,李明輝.求解股票期權(quán)定價(jià)問(wèn)題的有限差分方法[J].東北大學(xué)學(xué)報(bào),2004,25(2):190-193.
[6] 蹇明.BlackScholes期權(quán)定價(jià)模型的五點(diǎn)式混合差分方法[J].經(jīng)濟(jì)數(shù)學(xué),2011,28(4):66-70.
[7] K S CRUMP. Numerical inversion of Laplace transform using Fourier series approximation[J]. Assoc Comput Mach,1976,23(5):89-96.
[8] G HONIG, U HIRDES. A method for the numerical inversion of Laplace transform[J]. Journal of Computation and Applied Mathematics.1984,10(7):113-132.
[9] R MALLIER, G ALOBAIDI. Laplace transforms and American options[J].Applied Math Finance, 2000,7(4):241-256.
[10]F DURBIN. Numerical inversion of Laplace transforms:an efficient improvement to Dubner and Abate's method[J]. The Compute Journal.1974,17(4):371-376.
[11]F R D HOOG, J H KNIGHT, A N STOKES. An improved method for numerical inversion of Laplace transforms, SIAM [J]. Sci. Statist. Comput.1982,3(2):357-366.
[12]L HYOSEOP,S DONGWOO.Laplace transformation method for Black-Scholes[J].Internation Journal of Numerical Analysis and Modeling.2009,6(4):1-17.
[13]A TAGLIANI, M MILEV. Laplace transform and finite difference methods for BlackScholes equation[J].Applied Mathematics and Computation.2013,220(6):649-658.
[14]姜禮尚.期權(quán)定價(jià)的數(shù)學(xué)模型與方法(第二版)[M].北京:高等教育出版社,2008.
[15]約翰.赫爾.期權(quán)期貨及其他衍生產(chǎn)品(第8版)[M].北京,機(jī)械工業(yè)出版社,2012.
[4] 張鐵,祝丹梅.美式期權(quán)定價(jià)問(wèn)題的變網(wǎng)格差分方法[J].計(jì)算數(shù)學(xué),2008,30(4):379-387.
[5] 張鐵,李明輝.求解股票期權(quán)定價(jià)問(wèn)題的有限差分方法[J].東北大學(xué)學(xué)報(bào),2004,25(2):190-193.
[6] 蹇明.BlackScholes期權(quán)定價(jià)模型的五點(diǎn)式混合差分方法[J].經(jīng)濟(jì)數(shù)學(xué),2011,28(4):66-70.
[7] K S CRUMP. Numerical inversion of Laplace transform using Fourier series approximation[J]. Assoc Comput Mach,1976,23(5):89-96.
[8] G HONIG, U HIRDES. A method for the numerical inversion of Laplace transform[J]. Journal of Computation and Applied Mathematics.1984,10(7):113-132.
[9] R MALLIER, G ALOBAIDI. Laplace transforms and American options[J].Applied Math Finance, 2000,7(4):241-256.
[10]F DURBIN. Numerical inversion of Laplace transforms:an efficient improvement to Dubner and Abate's method[J]. The Compute Journal.1974,17(4):371-376.
[11]F R D HOOG, J H KNIGHT, A N STOKES. An improved method for numerical inversion of Laplace transforms, SIAM [J]. Sci. Statist. Comput.1982,3(2):357-366.
[12]L HYOSEOP,S DONGWOO.Laplace transformation method for Black-Scholes[J].Internation Journal of Numerical Analysis and Modeling.2009,6(4):1-17.
[13]A TAGLIANI, M MILEV. Laplace transform and finite difference methods for BlackScholes equation[J].Applied Mathematics and Computation.2013,220(6):649-658.
[14]姜禮尚.期權(quán)定價(jià)的數(shù)學(xué)模型與方法(第二版)[M].北京:高等教育出版社,2008.
[15]約翰.赫爾.期權(quán)期貨及其他衍生產(chǎn)品(第8版)[M].北京,機(jī)械工業(yè)出版社,2012.
[4] 張鐵,祝丹梅.美式期權(quán)定價(jià)問(wèn)題的變網(wǎng)格差分方法[J].計(jì)算數(shù)學(xué),2008,30(4):379-387.
[5] 張鐵,李明輝.求解股票期權(quán)定價(jià)問(wèn)題的有限差分方法[J].東北大學(xué)學(xué)報(bào),2004,25(2):190-193.
[6] 蹇明.BlackScholes期權(quán)定價(jià)模型的五點(diǎn)式混合差分方法[J].經(jīng)濟(jì)數(shù)學(xué),2011,28(4):66-70.
[7] K S CRUMP. Numerical inversion of Laplace transform using Fourier series approximation[J]. Assoc Comput Mach,1976,23(5):89-96.
[8] G HONIG, U HIRDES. A method for the numerical inversion of Laplace transform[J]. Journal of Computation and Applied Mathematics.1984,10(7):113-132.
[9] R MALLIER, G ALOBAIDI. Laplace transforms and American options[J].Applied Math Finance, 2000,7(4):241-256.
[10]F DURBIN. Numerical inversion of Laplace transforms:an efficient improvement to Dubner and Abate's method[J]. The Compute Journal.1974,17(4):371-376.
[11]F R D HOOG, J H KNIGHT, A N STOKES. An improved method for numerical inversion of Laplace transforms, SIAM [J]. Sci. Statist. Comput.1982,3(2):357-366.
[12]L HYOSEOP,S DONGWOO.Laplace transformation method for Black-Scholes[J].Internation Journal of Numerical Analysis and Modeling.2009,6(4):1-17.
[13]A TAGLIANI, M MILEV. Laplace transform and finite difference methods for BlackScholes equation[J].Applied Mathematics and Computation.2013,220(6):649-658.
[14]姜禮尚.期權(quán)定價(jià)的數(shù)學(xué)模型與方法(第二版)[M].北京:高等教育出版社,2008.
[15]約翰.赫爾.期權(quán)期貨及其他衍生產(chǎn)品(第8版)[M].北京,機(jī)械工業(yè)出版社,2012.