鄭來運
(寧夏大學 機械工程學院, 銀川 750021)
?
具有Markovian調(diào)制的隨機資本系統(tǒng)數(shù)值解的收斂性
鄭來運
(寧夏大學 機械工程學院, 銀川 750021)
根據(jù)Euler數(shù)值方法,給出了一類具有Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng)的數(shù)值解,并應(yīng)用It公式、Burkholder-Davis-Gundy不等式和Gronwall引理證明了數(shù)值解的收斂性,給出了數(shù)值解收斂于解析解的充分條件。
隨機資本系統(tǒng);Markovian調(diào)制;It公式
考慮如下具有Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng):
(1)
帶Markovian調(diào)制的系統(tǒng)具有很好的優(yōu)點,它能描述(模型化)動力系統(tǒng)中結(jié)構(gòu)的突然變化,如系統(tǒng)組成部分的失敗或修復(fù)、突然的環(huán)境變化、子系統(tǒng)間互聯(lián)的改變以及對不同的非線性部分的操作等[1],因此受到了廣泛關(guān)注[2-17]。由于一般很難或無法獲得該系統(tǒng)的解析解,近年來,很多學者更加關(guān)注隨機微分方程數(shù)值解的研究。例如:Yuan等[2]討論了具有Markovian調(diào)制的隨機微分方程 數(shù)值解的收斂性;Wang等[5]研究了帶 Poisson 跳和 Markovian調(diào)制的隨機時滯微分方程數(shù)值解的收斂性;Rathinasamy等[6]給出了帶多時滯和Markovian調(diào)制的線性隨機微分方程半隱式Euler法的均方穩(wěn)定性,最近又討論了具有Markovian調(diào)制的年齡相關(guān)隨機種群系統(tǒng)分裂步數(shù)值方法[7];Zhou等[8]證明了在局部 Lipschitz 條件下,帶 Markovian調(diào)制的中立型時滯隨機微分方程數(shù)值解的收斂性;Li等[9]討論了帶Markovian調(diào)制的隨機時滯微分方程數(shù)值解的收斂性和穩(wěn)定性,并研究了帶Markovian調(diào)制[10]以及帶跳和Markovian調(diào)制[11]的年齡相關(guān)隨機種群方程數(shù)值解的收斂性;Jiang等[12]討論了帶Markovian調(diào)制的隨機時滯積分微分方程分裂步向后Euler數(shù)值解的穩(wěn)定性;張啟敏等[13-14]研究了帶Markovian調(diào)制的年齡相關(guān)隨機種群系統(tǒng)數(shù)值解的漸漸穩(wěn)定性以及半馴服Euler法的指數(shù)穩(wěn)定性。對于投資模型問題,Markovian調(diào)制模型在金融經(jīng)濟學的多個重要領(lǐng)域均有應(yīng)用[15-16],但對于帶Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng),相關(guān)研究較少[17]。本文討論給定條件下帶Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng)數(shù)值逼近解的收斂性。
設(shè)(Ω,F,P)是一個完備概率空間,{Ft}t≥0是其上的一個濾子且滿足一般性條件(即單調(diào)增右連續(xù),且F0包含所有的P零測集)。設(shè)r(t),t>0為定義在(Ω,F,P)上取值于有限狀態(tài)S={1,2,…,N}的右連續(xù)Markovian鏈,其生成元Γ=(γij)N×N定義如下:
其中,Δ>0,γij≥0(i≠j)表示從狀態(tài)i到狀態(tài)j的轉(zhuǎn)移概率,且
(2)
其中,Kt=K(a,t)。
對于系統(tǒng)(1),取Δt=h為離散時間步長(時間增量),則其Euler逼近解迭代式為
(3)
初始值
(4)
假定系統(tǒng)(1)滿足如下條件:
1)μ(a,t)在Q上非負可測,γ(t)和A(t)在[0;T]上非負連續(xù),滿足
2)f(i,0)=0,g(i,0)=0,i∈S。
3) (Lipschitz條件)存在正常數(shù)Ki,對任意x,y∈H,i∈S,有
若上述條件成立,則方程(1)在(a,t)∈Q上存在唯一解K(a,t),證明方法與文獻[18]中方法類似,這里不再贅述。
引理2若條件(A)-(D)成立,則存在常數(shù)k≥2和C1>0,使得
證明過程類似于文獻[17],且可得
證明由式(4)可得
對|Qt|2應(yīng)用It公式,有
于是得
對?t∈[0,T],有
利用條件3)得
基于QoS綜合匹配的語義Web服務(wù)選擇方法過程中,兩個QoS屬性參數(shù)之間相關(guān)性的臨界距離L,本文將其設(shè)定為1。QoS語義匹配成功后,對相應(yīng)的QoS數(shù)值進行匹配。
(5)
應(yīng)用Burkholder-Davis-Gundy不等式,對某些正常數(shù)M1>0,有
(6)
于是由式(5)和(6)可得
應(yīng)用 Gronwall 引理,得
證明完畢。
引理4對任意的t∈[0,T],存在正常數(shù)C3和C4,使得
引理5若條件1)~3)成立,則存在常數(shù)C5使
證明對任意的t∈[0,T],存在正整數(shù)m,使得t∈[mh,(m+1)h),有
于是,
應(yīng)用Cauchy-Schwarz不等式和假設(shè)條件,有
應(yīng)用Doob不等式和引理3,有
證明完畢。
由引理1~5可證明在給定條件1)~4)下,具有Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng)的數(shù)值解收斂到其解析解。
定理1若條件1)~4)成立,則
證明由式(2)和(4),有
于是,對?t∈[0,T],有
應(yīng)用Burkholder-Davis-Gundy不等式,得
其中,k1和k2均為正常數(shù)。于是有
再利用Gronwall不等式,可得
即有
定理2若條件1)~4)成立,則數(shù)值逼近解(4)收斂于系統(tǒng)(1)的解析解,即滿足
本文討論了一種具有Markovian調(diào)制的役齡相關(guān)隨機資本系統(tǒng)數(shù)值解的收斂性。結(jié)果表明:在相應(yīng)條件下,系統(tǒng)的數(shù)值逼近解(4)收斂于其解析解。
[1]MARITON M.Jump Linear Systems in Automatic Control[M].New York:Marcel Dekker,1990.
[2]YUAN C G,MAO X R.Convergence of the Euler-Maruyama method for stochastic differential equations with Markovian switching[J].Mathematics and Computers in Simulation,2004,64:223-235.
[3]WANG Z,LIU Y,YU L,et al.Exponential stability of delayed recurrent neural networks with Markovian jumping parameters[J].Physics Letters A,2006,356:346-352.
[4]戴偉星,胡適耕.帶Markov調(diào)制的隨機微分延遲方程的穩(wěn)定性(英文)[J].數(shù)學研究與評論,2008,28:511-520.
[5]WANG L S,XUE H.Convergence of numerical solutions to stochastic differential delay equations with Markovian switching and Poisson jump[J].Applied Mathematics and Computation 2007,188:1161-1172.
[6]RATHINASAMY A,BALACHANDRAN K.Mean square stability of semi-implicit Euler method for linear stochastic differential equations with multiple delays and Markovian switching[J].Applied Mathematics and Computation,2008,206:968-979.
[7]RATHINASAMY A.Split-step-methods for stochastic age-dependent population equations with Markovian switching[J].Nonlinear Analysis:Real World Applications,2012,13:1334-1345.
[8]Zhou S B,Wu F K.Convergence of numerical solutions to neutral stochastic delay differential equations with Markovian switching[J].Computational and Applied Mathematics,2009,229:85-96.
[9]LI R H,HOU Y M.Convergence and stability of numerical solutions to SDDE with Markovian Switching[J].Applied Mathematics and Computation,2006,175:1080-1091.
[10]LI R H,LEUNG P K,PANG W K.Convergence of numerical solutions to stochastic age-dependent population equations with Markovian switching[J].Journal of Computational and Applied Mathematics,2009,233:1046-1055.
[11]LI R H,PANG W K,LEUNG P K.Convergence of numerical solutions to stochastic age-structured population equations with diffusions and Markovian switching[J].Applied Mathematics and Computation,2010,216:744-752.
[12]JIANG F,SHEN Y,HU J H.Stability of the split-step backward Euler scheme for stochastic delay integro-differential equations with Markovian switching[J].Communications in Nonlinear Science and Numerical Simulation,2011,16:814-821.
[13]MA W J,ZHANG Q M.,WANG Z P.Asymptotic stability of stochastic age-dependent population equations with Markovian switching[J].Applied Mathematics and Computation,2014,227:309-319.
[14]楊洪福,張啟敏.具有Markov調(diào)制的隨機年齡結(jié)構(gòu)種群系統(tǒng)半馴服Euler法的指數(shù)穩(wěn)定性[J].數(shù)學年刊A輯(中文版),2016,37:7-88.
[15]GUO X.Information and Option Pricing[J].Quantitative Finance,2001(1):38-44.
[16]ELLIOTT R J,HINZ J.Portfolio analysis,hidden Markov models and chart analysis by PF-diagrams[J].International Journal of Theoretical and Applied Finance,2002,5:385-399.
[17]ZHANG Q M,LIU Y T,LI X N.Strong convergence of split-step backward Euler method for stochastic age-dependent capital system with Markovian switching[J].Applied Mathematics and Computation,2014,235:439-453.
[18]ZHANG Q M,LIU W A,NIE Z K.Existence,uniqueness and exponential stability for stochastic age-dependent population[J].Applied Mathematics and Computation,2004,154:183-201.
[19]ANDERSON W J.Continuous-time Markov Chains[M].Berlin:Springer,1991.
(責任編輯陳艷)
Convergence of Solution for Stochastic Capital System with Markovian Switching
ZHENG Lai-yun
(School of Mechanical Engineering, Ningxia University, Yinchuan 750021, China)
The numerical solution of a class of stochastic age-dependent capital system with Markovian switching was given according to the Euler method in time discretization. Utilizing It’s formula, Gronwall lemma and Barkholder-Davis-Gundy inequality, some criteria were obtained for the convergence of the numerical solution.
stochastic capital system; Markovian switching; It’s formula
2016-05-08
寧夏自然科學基金資助項目(NZ14048)
鄭來運(1979—),女,寧夏人,講師,主要從事運籌學與控制理論的研究,E-mail: zhenglaiyun@126.com。
format:ZHENG Lai-yun.Convergence of Solution for Stochastic Capital System with Markovian Switching[J].Journal of Chongqing University of Technology(Natural Science),2016(10):156-162.
10.3969/j.issn.1674-8425(z).2016.10.025
O231
A
1674-8425(2016)10-0156-07
引用格式:鄭來運.具有Markovian調(diào)制的隨機資本系統(tǒng)數(shù)值解的收斂性[J].重慶理工大學學報(自然科學),2016(10):156-162.