Hindawi Publishing Corporation Mathematical Problems in Engineering Volume 2013, Article ID 168169, 9 pages http://dx.doi.org/10.1155/2013/168169
Research Article Exponential Stability of Stochastic Nonlinear Dynamical Price System with Delay Wenli Zhu,1 Xinfeng Ruan,1 Ye Qin,2 and Jie Zhuang3 1
School of Economic Mathematics, Southwestern University of Finance and Economics, Chengdu 611130, China Teaching Research Training Center in Xindu District of Chengdu, Chengdu 610500, China 3 Sichuan University of Science and Engineering Library, Zigong 643000, China 2
Correspondence should be addressed to Wenli Zhu;
[email protected] Received 30 January 2013; Revised 28 April 2013; Accepted 17 May 2013 Academic Editor: Wuquan Li Copyright Β© 2013 Wenli Zhu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Based on Lyapunov stability theory, ItΛo formula, stochastic analysis, and matrix theory, we study the exponential stability of the stochastic nonlinear dynamical price system. Using Taylorβs theorem, the stochastic nonlinear system with delay is reduced to an n-dimensional semilinear stochastic differential equation with delay. Some sufficient conditions of exponential stability and corollaries for such price system are established by virtue of Lyapunov function. The time delay upper limit is solved by using our theoretical results when the system is exponentially stable. Our theoretical results show that if the classical price Rayleigh equation is exponentially stable, so is its perturbed system with delay provided that both the time delay and the intensity of perturbations are small enough. Two examples are presented to illustrate our results.
1. Introduction Let us make the following assumptions. (H1.1) Demand for product is quadratic function with respect to price. (H1.2) The price is not very sensitive to the change of inventory. That is, damping of nonlinear dynamical system π is a π-order infinitesimal (π > 0 is small enough). (H1.3) Stochastic noise is related to price. That is, it can be treated as Gaussian white noise, and the excitation coefficient is βπ-order infinitesimal. The price system can be described by linear equations because of their convenience in mathematical treatment. Therefore linear equations play an important role in theory and their applications. However, they can not perfectly describe the process of the price fluctuation in nonlinear version. Then the nonlinear equations should be employed, for their virtues that can deeply reflect the rules of price fluctuation. Suppose π(π‘), π·(π‘), π(π‘), and π(π‘) as supply, demand, price, and inventory at time π‘, respectively. π0 is initial
supply, π, π, π, π, πΌ and πΎ are constants, and alphabet with a bar is equilibrium value. Reference [1] gives a deterministic nonlinear price model as follows: ππ π2 π = π (ππ2 + ππ + π) β ππΌπ β π (π0 β π·) . 2 ππ‘ ππ‘
(1)
If there exists an equilibrium point in the above price system, denoted by π, then π = (π· β π0 )/πΌ and π· = π0 + πΌπ. Let π₯(π‘) = π(π‘) β π and π‘ = π/βπΌπ, π = βπ/πΌ. Applying Lienard transformation, we obtain the classical price Rayleigh equation as follows: 1 ππ₯ 1 = βπ¦ + ππΜ ( ππ₯3 + π0 π₯2 + π0 π₯) , ππ 3 2 ππ¦ = π₯, ππ 2
(2)
Μ where π0 = 2ππ + π, π0 = ππ + ππ + π, π β ππ. For many real-world systems, there always exist random disturbances such as the measurement error and the control input of the system [2β5]. The basic source of random
2
Mathematical Problems in Engineering
disturbance is Gaussian white noise, which represents the joint effects of a large number of independent random forces acting on the systems, and the influence of individual is not significant. By (H1.3), the stochastic nonlinear dynamical price system can be described by stochastic differential equation (SDE for short) as follows [5]: 1 1 ππ₯ = βππΌ [βπ¦ + ππΜ ( ππ₯3 + π0 π₯2 + π0 π₯)] ππ‘ 3 2 + π1/2 πΎ ππ (π‘) ,
(3)
ππ¦ = βππΌπ₯ ππ‘, where {π(π‘), π‘ β₯ 0} is 1-dimensional Brownian motion. The above system also can be rewritten as the following matrix form: ππ§ (π‘) = π (π‘, π§ (π‘)) ππ‘ + β (π‘, π§ (π‘)) ππ (π‘) ,
(4)
where π : π
+ Γ π
2 β π
2 , β : π
+ Γ π
2 β π
2 , π(π‘, π§(π‘)) = βππΌ(βπ¦+ππ(ππ₯ Μ 3 /3+π0 π₯2 /2+π0 π₯), π₯)π , π§(π‘) = (π₯(π‘), π¦(π‘))π β 2 π
, and β(π‘, π§(π‘)) = (βππΎ, 0)π . Supply is not only influenced by price and demand but also influenced by production management, information feedback, transportation, and so forth. Therefore, π(π‘) not only depends on the situation at π‘ but also on the certain period π‘ β π (π > 0 is a given time delay) in the past [6β 9]. Furthermore, the parameter perturbation of the systemβs internal structure should also be taken into account in this paper. Based on the abovementioned, the price system (4) can be extended to more general n-dimensional stochastic nonlinear price systems with delay as follows: ππ₯ (π‘) = [π (π‘, π₯ (π‘) , π₯ (π‘ β π)) + π (π‘, π₯ (π‘) , π₯ (π‘ β π))] ππ‘ + β (π‘, π₯ (π‘) , π₯ (π‘ β π)) ππ (π‘) , π₯ (π‘) = π (π‘) ,
π‘ β₯ 0,
βπ β€ π‘ β€ 0, (5)
where π is a given time delay. The maps π, π β πΆ (π
+ Γ π
π Γ π
π , π
π ), β β πΆ (π
+ Γ π
π Γ π
π , π
πΓπ ). π represents the uncertainty. {π(π‘), π‘ β₯ 0} is an m-dimensional Brownian motion, and the term β (π‘, π₯(π‘), π₯(π‘ β π)) ππ(π‘) represents the stochastic disturbance. Furthermore, we always assume that π(π‘, 0, 0) = π(π‘, 0, 0) = β(π‘, 0, 0) β‘ 0 for the stability purpose of this paper. Stability is a very important dynamical feature for the stochastic price system with delay, and it is one of the main purposes of system designing [5, 6]. Keeping the price system steady within the cycle as long as possible to avoid inflation or deflation has the vital significance for the healthy development of the economy of the country. There is a rich literature on time delay system and stochastic system. Stability of stochastic system has been studied. See, for example, Liu and Feng [2], Liu and Deng [3], Yong and Zhou [4], Li and Xu [5], and Mao [10]. Stability of time delay system has been studied. See, for example, Kazmerchuk et al.
[6], Lv and Liu [7], Lv and Zhou [8], Zhu et al. [11], Zhu and Yi [12], and Trinh and Aldeen [13]. Mao [14], Mao and Shah [15], Zhu and Hu [16], Zhu and Hu [17], and S. Xie and L. Xie [18] established some stability criteria of the stochastic system with delay by using an LMI approach. The Hopf bifurcation of price Rayleigh delayed equation on deterministic case has been studied extensively in recent years. See, for example, [6β8]. The stability and the optimal control of stochastic nonlinear dynamical price model has been studied in [5]. Unfortunately, there is a little previous literature on stochastic nonlinear dynamical price system with delay. Thus, we aim to fill this gap in this paper. We plug the time delay, the parameter perturbation, and the stochastic item into nonlinear dynamical price system (2). Such models may be identified as stochastic differential delayed equations (SDDEs for short). Our target in this paper is to derive some sufficient conditions of exponential stability for SDDEs. Li and Xu only analyzed the stability for SDEs (4) in virtue of the marginal probability density about π₯ in [5] but did not give the sufficient condition for stability. In this paper, using Taylorβs theorem, the n-dimensional nonlinear SDDE (5) is reduced to an n-dimensional semilinear SDDE correspondingly. Some sufficient conditions of exponential stability and corollaries for such price system are established by using Lyapunov function. The time delay upper limit is solved by using our theoretical results when the system is exponentially stable. Thus, [5] is promoted and improved. Our theoretical results show that if the classical price Rayleigh equation (2) is exponentially stable, so is its perturbed system (5) with delay provided that both the time delay and the intensity of perturbations are small enough. Those results will help our government make a macrocontrol for price system and timely adjust their pricing strategies. The rest of this paper is organized as follows. In Section 2, we introduce the definition of the exponential stability of SDDEs. Section 3 is devoted to the sufficient conditions for exponential stability and almost surely exponential stability of price system. Section 4 presents two simple examples to illustrate our results. Finally, Section 5 concludes the paper.
2. Preliminaries Throughout this paper and unless specified, we let π(π‘) = (π1 (π‘), . . . , ππ (π‘))π be an m-dimensional Brownian motion defined on a complete probability space (Ξ©, F, π) with a natural filtration {Fπ‘ }π‘β₯0 (i.e., Fπ‘ = π{π(π ) : 0 β€ π β€ π‘} and augmented by all the π-null sets in F). Denote by | β
| the Euclidean norm. If π΄ is a vector or matrix, its transpose is denoted by π΄π . If π΄ is a matrix, denote by βπ΄β the operator norm of π΄, that is, βπ΄β = sup{|π΄π₯| : |π₯| = 1}. π(β
) β πΆ[βπ, 0] is the initial path of π₯, where π > 0 is a given finite time delay and πΆ[βπ, 0] is the set of continuous functions from [βπ, 0] π into π
π . Moreover, denote by πΏ F0 (βπ, 0; π
π ) the family of π
π valued adapted stochastic processes π(π ), βπ β€ π β€ 0 such 0 that π(π ) is F0 -measurable and β«βπ πΈ|π(π )|π ππ < +β (π > 1). We define a norm mean square in ππ as follows: 1/2
|π₯|MS = (πΈ|π₯|2 )
,
for any π₯ β ππ ,
(6)
Mathematical Problems in Engineering
3
where ππ is the set of random variable in probability space (Ξ©, F, π). Similarly, we define βπ΄βMS = sup{|π΄π₯|MS : |π₯|MS = 1}. Then, the ItΛo integral of β(π‘, π₯(π‘)) (from π to π) is defined by π
Definition 3. The trivial solution of system (5) is said to be almost surely exponentially stable, if there exists a positive constant π such that 1 σ΅¨ σ΅¨ lim sup ln σ΅¨σ΅¨σ΅¨π₯ (π‘; π)σ΅¨σ΅¨σ΅¨ β€ βπ π‘ββ π‘
a.s.
(10)
π
for any π β πΏ F0 (βπ, 0; π
π ), where βπ is called almost surely Lyapunov exponent of the trivial solution.
(πΌ) β« β (π‘, π₯ (π‘)) ππ (π‘) π
π
= limMS β β (π‘π , π₯ (π‘π )) [π (π‘π+1 ) β π (π‘π )]
(7)
πβ0 π=0
(limit in πΏ2 (π)) , πΓπ
and where β is a stochastic process with value in π π 2 ππ‘ < β (0 β€ π < π). Let π = π‘0 < π‘1 < β«π ββ(π‘, π₯(π‘))βMS β
β
β
< π‘π = π, π = max{Ξπ‘π , π = 0, 1, . . . , π β 1}, Ξπ‘π = π‘π+1 β π‘π , and limMS be a limit in mean square sense. It is proved directly from the definition of ItΛo integral that π
1 1 β« π (π‘) ππ (π‘) = [π2 (π) β π2 (π)] β (π β π) , 2 2 π
To obtain sufficient condition of the exponential stability of system (5), we assume that the functions π(π‘, π₯, π¦), π(π‘, π₯, π¦) are 1-order continuously differentiable in the neighbourhood of (π‘, 0, 0) with respect to (π₯, π¦) β π
π Γπ
π . According to Taylor expansion, for 0 < π < 1 π (π‘, π₯, π¦) = ππ₯ (π‘, ππ₯, ππ¦) π₯ + ππ¦ (π‘, ππ₯, ππ¦) π¦, π (π‘, π₯, π¦) = ππ₯ (π‘, ππ₯, ππ¦) π₯ + ππ¦ (π‘, ππ₯, ππ¦) π¦.
(8)
(11)
Thus,
where π(β
) is 1-dimensional Brownian motion. The extra term β(π β π)/2 shows that the ItΛo stochastic integral does not behave like ordinary integrals. It leads to ItΛo-type stochastic system being different from non-ItΛo-type. See [19] for the details. Now, let us present an existence and uniqueness result for system (5). First, we make the following assumptions for the coefficients of (5). (H2.1) The maps π, π, and β are locally Lipschitz continuous. (H2.2) The maps π, π, and β satisfy the linear growth condition. Theorem 1 (see [4]). Let (H2.1) and (H2.2) hold. Then, for any π(π‘) β πΏ2F0 (βπ, 0; π
π ), (5) has a unique strong solution which is denoted by π₯(π‘; π), and π₯(π‘; π) is square integrable. So, (5) has a trivial solution π₯(π‘; 0) = 0. For stochastic system, exponential stability in mean square and almost surely exponential stability are generally used [2]. Definition 2. The trivial solution of system (5) is said to be pth moment exponentially stable, if there exists a positive constant π such that 1 σ΅¨π σ΅¨ lim sup ln (πΈσ΅¨σ΅¨σ΅¨π₯ (π‘; π)σ΅¨σ΅¨σ΅¨ ) β€ βπ π‘ββ π‘
3. Exponential Stability for Stochastic Price System with Delay
(9)
π
for any π β πΏ F0 (βπ, 0; π
π ), where βπ is called pth moment Lyapunov exponent of the trivial solution. In particular, π = 2; it is called mean square exponentially stable.
π (π‘, π₯, π¦) + π (π‘, π₯, π¦) = (ππ₯ (π‘, ππ₯, ππ¦) + ππ₯ (π‘, ππ₯, ππ¦)) π₯
(12)
+ (ππ¦ (π‘, ππ₯, ππ¦) + ππ¦ (π‘, ππ₯, ππ¦)) π¦. Therefore, system (5) can be reduced to an n-dimensional semilinear stochastic differential delayed equation as follows: ππ₯ (π‘) = [π (π‘, π₯ (π‘) , π₯ (π‘ β π)) + π (π‘, π₯ (π‘) , π₯ (π‘ β π))] ππ‘ + β (π‘, π₯ (π‘) , π₯ (π‘ β π)) ππ (π‘) , π₯ (π‘) = π (π‘) ,
π‘ β₯ 0,
βπ β€ π‘ β€ 0, (13)
where π΄ 1 , π΅1 are π Γ π matrices. π΄ 1 (π‘), π΄ 2 (π‘ β π), π΅1 (π‘), and π΅2 (π‘ β π) represent the uncertainties. They are bounded π Γ π matrix-valued functions. Here, π, π(π‘), β(π‘, π₯(π‘), π₯(π‘ β π)), and π(π‘) are the same as in the previous section. We make the following assumption for the coefficients of system (13). (H3.1) There exist nonnegative constants πΌπ , π½π (π = 1, 2, 3), for any π‘ β₯ 0 such that σ΅© σ΅©σ΅© σ΅©σ΅©π΄ 1 (π‘)σ΅©σ΅©σ΅© β€ πΌ1 , σ΅©σ΅© σ΅© σ΅©σ΅©π΅1 (π‘)σ΅©σ΅©σ΅© β€ π½1 ,
σ΅© σ΅©σ΅© σ΅©σ΅©π΄ 2 (π‘ β π)σ΅©σ΅©σ΅© β€ πΌ2 , σ΅© σ΅©σ΅© σ΅©σ΅©π΅2 (π‘ β π)σ΅©σ΅©σ΅© β€ π½2
(14)
and for any (π‘, π₯(π‘), π₯(π‘ β π)) β π
+ Γ π
π Γ π
π such that tr [βπ (π‘, π₯ (π‘) , π₯ (π‘ β π)) β (π‘, π₯ (π‘) , π₯ (π‘ β π))] β€ πΌ3 |π₯ (π‘)|2 + π½3 |π₯ (π‘ β π)|2 .
(15)
4
Mathematical Problems in Engineering
Theorem 4. Let β satisfy (H2.1)-(H2.2), and condition (14) hold. Then, for any π(π‘) β πΏ2F0 (βπ, 0; π
π ), system (13) has a unique strong solution which is denoted by π₯(π‘; π), and π₯(π‘; π) is square integrable. So, (13) has a trivial solution π₯(π‘; 0) = 0. See Mao [10] for the proof of Theorem 4. In the study of mean square exponential stability, it is often to use a quadratic function as the Lyapunov function; that is, π(π‘, π₯) = π₯π (π‘)πΊπ₯(π‘), where πΊ is a symmetric positive definite π Γ π matrix (see [11, 20]). Theorem 5. Let (H3.1) holds, and then the trivial solution of system (13) is exponentially stable in the mean square. Assume that there exists a pair of symmetric positive definite π Γ π matrices πΊ and π such that πΊ (π΄ + π΅) + (π΄ + π΅)π πΊ = βπ,
(16)
By (14), we can estimate the second item and the third item of (18), respectively, 2π₯π (π‘) πΊ [π΄ 1 (π‘) π₯ (π‘) + π΅1 (π‘) π₯ (π‘ β π)] β€ βπΊβ [(2πΌ1 + π½1 ) |π₯ (π‘)|2 + π½1 |π₯ (π‘ β π)|2 ] , (21)
π
2π₯ (π‘) πΊ [π΄ 2 (π‘ β π) π₯ (π‘) + π΅2 (π‘) π₯ (π‘ β π)] β€ βπΊβ [(2πΌ2 + π½2 ) |π₯ (π‘)|2 + π½2 |π₯ (π‘ β π)|2 ] . By (15), the last item of (18) yields tr [βπ (π‘, π₯ (π‘) , π₯ (π‘ β π)) πΊβ (π‘, π₯ (π‘) , π₯ (π‘ β π))] (22)
β€ βπΊβ [πΌ3 |π₯ (π‘)|2 + π½3 |π₯ (π‘ β π)|2 ] .
π min (π) > βπΊβ (2πΌ1 + 2πΌ2 + 2π½1 + 2π½2 + πΌ3 + π½3 ) + 2 βπΊπ΅β
Substituting (21) and (22) into (18), we get
β
β2π [6π (βπ΄β2 + βπ΅β2 + πΌ12 + π½12 + πΌ22 + π½22 ) + πΌ3 + π½3 ], (17) where π min (π) > 0 is the smallest eigenvalue of π.
π [π₯π (π‘) πΊπ₯ (π‘)] β€ β [π min (π) β βπΊβ (2πΌ1 + π½1 + 2πΌ2 + π½2 + πΌ3 ) β π½]
Proof. Fix the initial data π(π‘) arbitrarily, and write π₯(π‘; π) = π₯(π‘) simply. Applying ItΛoβs formula to π₯π (π‘)πΊπ₯(π‘), we have
β
|π₯ (π‘)|2 ππ‘ + βπΊβ (π½1 + π½2 + π½3 ) |π₯ (π‘ β π)|2 ππ‘
π [π₯π (π‘) πΊπ₯ (π‘)]
+
= 2π₯π (π‘) πΊ [π΄π₯ (π‘) + π΅π₯ (π‘ β π)] ππ‘
+ 2π₯π (π‘) πΊβ (π‘, π₯ (π‘) , π₯ (π‘ β π)) ππ (π‘) .
+ 2π₯π (π‘) πΊ [π΄ 1 (π‘) π₯ (π‘) + π΅1 (π‘) π₯ (π‘ β π)] ππ‘
(23)
+ 2π₯π (π‘) πΊ [π΄ 2 (π‘ β π) π₯ (π‘) + π΅2 (π‘ β π) π₯ (π‘ β π)] ππ‘ π
+ 2π₯ πΊβ (π‘, π₯ (π‘) , π₯ (π‘ β π)) ππ (π‘) + tr [βπ (π‘, π₯ (π‘) , π₯ (π‘ β π)) πΊβ (π‘, π₯ (π‘) , π₯ (π‘ β π))] ππ‘. (18) By (16), we can estimate the first item of (18) as follows:
= 2π₯π (π‘) πΊ [π΄π₯ (π‘) + π΅π₯ (π‘ β π)] + 2π₯π (π‘) πΊπ΅π₯ (π‘) β 2π₯π (π‘) πΊπ΅π₯ (π‘)
π min (π) = βπΊβ (2πΌ1 + 2πΌ2 + π½1 + π½2 + πΌ3 + π)
+ (19)
2
β€ βπ min (π) |π₯ (π‘)| + π½|π₯ (π‘)| +
If (17) holds, then we can choose π > 0 small enough such that
+ βπΊβ (π½1 + π½2 + π½3 ) πππ + π½
2π₯π (π‘) πΊ [π΄π₯ (π‘) + π΅π₯ (π‘ β π)]
2
1 βπΊπ΅β2 β
|π₯ (π‘) β π₯ (π‘ β π)|2 ππ‘ π½
1 βπΊπ΅β2 β
|π₯ (π‘) β π₯ (π‘ β π)|2 , π½
where π½ = βπΊπ΅β β
β2π [6π (βπ΄β2 + βπ΅β2 + πΌ12 + π½12 + πΌ22 + π½22 ) + πΌ3 + π½3 ]. (20)
1 βπΊπ΅β2 {2π [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] πππ π½ + 2π [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] π2ππ } . (24)
Applying ItΛoβs formula to πππ‘ π₯π (π‘) πΊπ₯(π‘), we have π [πππ‘ π₯π (π‘) πΊπ₯ (π‘)] = ππππ‘ π₯π (π‘) πΊπ₯ (π‘) ππ‘ + πππ‘ π [π₯π (π‘) πΊπ₯ (π‘)] ,
for any π‘ β₯ 0. (25)
Mathematical Problems in Engineering
5 Next, recalling (14) and (15), for π β₯ π, we derive
Substituting (23) into (25) yields πππ‘ π₯π (π‘) πΊπ₯ (π‘)
πΈ|π₯ (π ) β π₯ (π β π)|2 π‘
π
β€ ππ (0) πΊπ (0) + π β« πππ π₯π (π ) πΊπ₯ (π ) ππ
β€ 2π β
πΈ β«
0
π βπ
6 [(βπ΄β2 + πΌ12 + πΌ22 ) |π₯ (π‘)|2
π‘
+ (βπ΅β2 + π½12 + π½22 ) |π₯ (π‘ β π)|2 ] ππ‘
β β« πππ [π min (π) β βπΊβ (2πΌ1 + 2πΌ2 + π½1 + π½2 + πΌ3 ) β π½] 0
π‘
2
ππ
+ 2πΈ β«
2
β
|π₯ (π )| ππ + β« π β
βπΊβ (π½1 + π½2 + π½3 ) |π₯ (π β π)| ππ
π βπ
0
π‘
ππ
(πΌ3 |π₯ (π‘)|2 + π½3 π₯(π‘ β π)2 ) ππ‘
β€ 2 [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] β«
1 + β« π β
βπΊπ΅β2 β
|π₯ (π ) β π₯ (π β π)|2 ππ π½ 0 ππ
π‘
π
π
π βπ
+ 2 [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] β«
π
+ β« π β
2π₯ (π ) πΊβ (π , π₯ (π ) , π₯ (π β π)) ππ (π ) . 0
πΈ|π₯ (π‘)|2 ππ‘
π
π βπ
πΈ|π₯ (π‘ β π)|2 ππ‘. (29)
(26) Similar to (28), for π‘ β₯ π, we have
Taking the expectation in (26), we have
π‘
πΈ (πππ‘ π₯π (π‘) πΊπ₯ (π‘))
β« πππ β
πΈ|π₯ (π ) β π₯ (π β π)|2 ππ 0
β€ πΈ (ππ (0) πΊπ (0))
π
= β« πππ β
πΈ|π₯ (π ) β π₯ (π β π)|2 ππ
β [π min (π) β βπΊβ (2πΌ1 + 2πΌ2 + π½1 + π½2 + πΌ3 + π) β π½]
0
π‘
π‘
+ β« πππ β
πΈ|π₯ (π ) β π₯ (π β π)|2 ππ
β
β« πππ β
πΈ|π₯ (π )|2 ππ
π
0
β€ π2 + 2 [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ]
π‘
+ βπΊβ (π½1 + π½2 + π½3 ) β« πππ β
πΈ|π₯ (π β π)|2 ππ 0
π‘ 1 + βπΊπ΅β2 β« πππ β
πΈ|π₯ (π ) β π₯ (π β π)|2 ππ . π½ 0
In order to give an estimate of πΈ(πππ‘ π₯π (π‘)πΊπ₯(π‘)), we now estimate the last two terms on the right-hand side of (27). First of all, for any π‘ β₯ π, we have π‘
β« πππ β
πΈ|π₯ (π β π)|2 ππ 0
ππ
2
π‘
ππ
π
π βπ
2
π‘
π
π
π βπ
β
β« πππ β«
π‘
π
π
π βπ
0
= π1 πππ + πππ β
β«
π‘βπ
0
0
πππ ππ ) π]
0
π‘
π
π
π βπ
β« πππ β«
(π’ = π β π)
(31)
πΈ|π₯ (] β π)|2 π] ππ
π‘
(]+π)β§π‘
0
]β¨π
= β« πΈ|π₯ (] β π)|2 (β«
πππ β
πΈ|π₯ (π )|2 ππ ,
where π1 = β«βπ πΈ|π₯(π )| ππ .
]β¨π
Similarly,
(28) 2
0
π‘
0
πππ’ β
πΈ|π₯ (π’)|2 ππ’
(]+π)β§π‘
β€ ππππ β« ππ] β
πΈ|π₯ (])|2 π].
π‘
σ΅¨2 σ΅¨ = πππ β
β« πΈσ΅¨σ΅¨σ΅¨π (π’)σ΅¨σ΅¨σ΅¨ ππ’ + πππ βπ π‘βπ
π‘
πΈ|π₯ (])|2 π] ππ = β« πΈ|π₯ (])|2 (β«
σ΅¨2 σ΅¨ β€ πππ β
β« πΈσ΅¨σ΅¨σ΅¨π(π β π)σ΅¨σ΅¨σ΅¨ ππ + πππ β
β« ππ(π βπ) β
πΈ|π₯ (π β π)|2 ππ 0 π
β
β«
πΈ|π₯ (] β π)|2 π] ππ ,
π
π
π
πΈ|π₯ (])|2 π] ππ
where π2 = β«0 πππ β
πΈ|π₯(π ) β π₯(π β π)|2 ππ . Moreover, β« πππ β«
= β« π β
πΈ|π₯ (π β π)| ππ + β« π β
πΈ|π₯ (π β π)| ππ 0
π
+ 2 [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ]
(27)
π
π‘
β
β« πππ β«
(30)
π‘
πππ ππ ) π]
β€ ππππ β« ππ] β
πΈ|π₯ (] β π)|2 π]. 0
(32)
6
Mathematical Problems in Engineering Since πΊ is positive definite,
Substituting (28) into (32) yields π‘
π
π
π βπ
β« πππ β«
πΈ|π₯ (] β π)|2 π] ππ
β€ ππππ (π1 πππ + πππ β
β«
π‘βπ
0
π₯π (π‘) πΊπ₯ (π‘) β₯ π min (πΊ) |π₯ (π‘)|2 , πππ β
πΈ|π₯ (π )|2 ππ )
(33)
π‘
< ππ1 π2ππ + ππ2ππ β« πππ β
πΈ|π₯ (π )|2 ππ . 0
where π min (πΊ) > 0 is the smallest eigenvalue of πΊ. Thus, πΈ (πππ‘ π₯π (π‘) πΊπ₯ (π‘)) β₯ πΈ (πππ‘ π min (πΊ) |π₯ (π‘)|2 ) .
Substituting (31) and (33) into (30), for π‘ β₯ π, we get π‘
β« πππ β
πΈ|π₯ (π ) β π₯ (π β π)|2 ππ
(37)
(38)
It then follows from (35) that
0
β€ π2 + 2 [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ]
πΈ|π₯ (π‘)|2 β€
π‘
β
ππππ β« ππ] β
πΈ|π₯ (])|2 π]
(39)
Hence, 2
+ 2 [6π (βπ΅β + β
(ππ1 π
for any π‘ β₯ π.
(34)
0
2ππ
π3 β
πβππ‘ π min (πΊ)
π½12
2ππ
+ ππ
+ π‘
π½22 )
+ π½3 ]
ππ
π3 1 1 β
πβππ‘ ) ln (πΈ|π₯ (π‘)|2 ) β€ ln ( π‘ π‘ π min (πΊ)
2
β« π β
πΈ|π₯ (π )| ππ ) . 0
π3 1 ). = β π + ln ( π‘ π min (πΊ)
Substituting (28) and (34) into (27) and recalling (24), we obtain that for π‘ β₯ π πΈ (πππ‘ π₯π (π‘) πΊπ₯ (π‘))
(40)
This easily yields
β€ πΈ (ππ (0) πΊπ (0)) 1 lim sup ln (πΈ|π₯ (π‘)|2 ) β€ βπ. π‘ββ π‘
β [π min (π) β βπΊβ (2πΌ1 + 2πΌ2 + π½1 + π½2 + πΌ3 + π) β π½]
(41)
π‘
β
β« πππ β
πΈ|π₯(π )|2 ππ 0
+ βπΊβ (π½1 + π½2 + π½3 ) (π1 πππ + πππ β
β«
π‘βπ
0
+
πππ β
πΈ|π₯ (π )|2 ππ )
1 βπΊπ΅β2 {π2 + 2 [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] π½
Remark 6. In the proof we gave, the estimate for the second moment Lyapunov exponent should not be greater than βπ. Theorem 7. The trivial solution of system (13) is also almost surely exponentially stable under the same assumption as Theorem 5.
π‘
β
ππππ β« πππ β
πΈ|π₯ (π )|2 ππ 0
Proof. For π‘ β [ππ, (π + 1)π], π = 2, 3, . . ., we have
+ 2 [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] π‘
β
(ππ1 π2ππ + ππ2ππ β« πππ β
πΈ|π₯ (π )|2 ππ )} 0
|π₯ (π‘)|2 β€ 3|π₯ (ππ)|2
= π3 , (35) where π3 = πΈ (ππ (0) πΊπ (0)) + βπΊβ (π½1 + π½2 + π½3 ) β
π1 πππ +
Then, (13) is exponentially stable in the mean square. The proof is complete.
1 βπΊπ΅β2 {π2 + 2 [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] β
ππ1 π2ππ } . π½ (36)
σ΅¨σ΅¨ (π+1)π σ΅¨ + 3 σ΅¨σ΅¨σ΅¨σ΅¨β« [(π΄ + π΄ 1 (π‘) + π΄ 2 (π‘ β π)) π₯ (π‘) σ΅¨σ΅¨ ππ σ΅¨σ΅¨2 σ΅¨ + (π΅ + π΅1 (π‘) + π΅2 (π‘ β π)) π₯ (π‘ β π)] ππ‘σ΅¨σ΅¨σ΅¨σ΅¨ σ΅¨σ΅¨ σ΅¨σ΅¨ (π+1)π σ΅¨σ΅¨2 σ΅¨ σ΅¨ + 3σ΅¨σ΅¨σ΅¨σ΅¨β« β (π‘, π₯ (π‘) , π₯ (π‘ β π)) ππ (π‘)σ΅¨σ΅¨σ΅¨σ΅¨ . σ΅¨σ΅¨ ππ σ΅¨σ΅¨ (42)
Mathematical Problems in Engineering
7 Let π0 β (0, π) be arbitrary. By Doobβs martingale inequality, it follows from (45) that
Recalling (14) and (15), we derive πΈ(
sup
ππβ€π‘β€(π+1)π
|π₯ (π‘)|2 )
β€ 3πΈ|π₯ (ππ)| + 18π β«
π (π :
2
(π+1)π
ππ
2
[ (βπ΄β +
πΌ12
+
(π+1)π
ππ
sup
ππβ€π‘β€(π+1)π
ln |π₯ (π‘)| β€ β
π3 β
πβπππ π min (πΊ)
π3 πΈ|π₯ (π‘ β π)|2 β€ β
πβπ(π‘βπ) π min (πΊ)
ππβ€π‘β€(π+1)π
β€
for π‘ β π β₯ π.
(π+1)π
π 1 lim sup ln |π₯ (π‘)| β€ β 2 π‘ββ π‘
πβππ‘ ππ‘ + [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ]
(π+1)π
ππ
a.s.
(50)
a.s.
(51)
Then, (13) is almost surely exponentially stable. The proof is complete. Remark 8. Again in the proof we gave, the estimate for the almost surely Lyapunov exponent should not be greater than βπ/2.
πβππ‘ ππ‘}
3π3 {πβπππ + ( [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] π min (πΊ)
Let us single out three important special cases. Case 1. If β(π‘, π₯(π‘), π₯(π‘ β π)) β‘ 0, then (13) reduces to a semilinear differential delay equation
+ [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] πππ ) 1 βπππ (1 β πβππ )} π π
ππ₯ (π‘) = [(π΄ + π΄ 1 (π‘) + π΄ 2 (π‘ β π)) π₯ (π‘)
3π3 β€ {πβπππ +([6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] π min (πΊ)
+ (π΅ + π΅1 (π‘) + π΅2 (π‘ β π)) π₯ (π‘ β π)] ππ‘.
+ [6π(βπ΅β2 + π½12 + π½22 ) + π½3 ]πππ ) ππβπππ } =
(49)
a.s.
π β π0 1 lim sup ln |π₯ (π‘)| β€ β 2 π‘ββ π‘
3π3 {πβπππ + [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] π min (πΊ)
β
(48)
Since π0 is arbitrary, we must have
ππ
β€
a.s.
This easily yields
|π₯ (π‘)| )
β
πππ β«
(π β π0 ) ππ 2
π β π0 1 ln |π₯ (π‘)| β€ β π‘ 2
(44)
2
β
β«
(47)
Hence,
for ππ β₯ π,
Substituting the above two into (43) yields sup
|π₯ (π‘)| β€ πβ(πβπ0 )ππ/2
(43)
πΈ|π₯ (ππ)|2 β€
(46)
holds in probability 1. Thus, for ππ β€ π‘ β€ (π + 1)π, and π β₯ π0 (π), we have
[πΌ3 πΈ|π₯ (π‘)|2 + π½3 πΈ|π₯ (π‘ β π)|2 ] ππ‘.
By (39), we easily get
πΈ(
ππβ€π‘β€(π+1)π
|π₯ (π‘)| > πβ(πβπ0 )ππ/2 ) β€ π4 πβπ0 ππ .
It then follows from the Borel-Cantelli lemma that for almost all π β Ξ©, there exists a π0 (π), π β₯ π0 (π), and
πΌ22 ) πΈ|π₯ (π‘)|2
+ (βπ΅β2 + π½12 + π½22 ) πΈ|π₯ (π‘ β π)|2 ] ππ‘ + 3β«
sup
3ππ3 1 { + [6π (βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] π min (πΊ) π
(52)
Corollary 9. Let condition (14) holds. Assume that there exists a pair of symmetric positive definite πΓπ matrices πΊ and π such that πΊ (π΄ + π΅) + (π΄ + π΅)π πΊ = βπ,
(53)
π min (π)
+ [6π (βπ΅β2 + π½12 + π½22 ) + π½3 ] πππ } β
πβπππ
> 2 βπΊβ (πΌ1 + πΌ2 + π½1 + π½2 )
= π4 πβπππ , (45) where π4 = (3ππ3 /π min (πΊ)){(1/π) + [6π(βπ΄β2 + πΌ12 + πΌ22 ) + πΌ3 ] + [6π(βπ΅β2 + π½12 + π½22 ) + π½3 ]πππ }.
+ 4π βπΊπ΅β β
β3 (βπ΄β2 + βπ΅β2 + πΌ12 + π½12 + πΌ22 + π½22 ). (54) Then, (52) is exponentially stable in the mean square.
8
Mathematical Problems in Engineering
Case 2. Let us further assume that π΄ 1 (π‘) = π΄ 2 (π‘ β π) = π΅1 (π‘) = π΅2 (π‘ β π) β‘ 0, then (52) becomes an ordinary differential equation
For convenience, let us choose π = the 2-order identity matrix, and then π min (π) = 1. By plugging these into (16), it is easy to find that
ππ₯ (π‘) = (π΄π₯ (π‘) + π΅π₯ (π‘ β π)) ππ‘.
1 [6 0] πΊ = [ 1] . 0 [ 4]
(55)
Corollary 10. Assume that there exists a pair of symmetric positive definite π Γ π matrices πΊ and π such that
(62)
We obtain via a simple calculation π
πΊ (π΄ + π΅) + (π΄ + π΅) πΊ = βπ, 2
(56) 2
π min (π) > 4π βπΊπ΅β β
β3 (βπ΄β + βπ΅β ).
(57)
Then, (55) is exponentially stable in the mean square. Remark 11. Corollary 10 clearly shows that if the price system Μ π₯(π‘) = (π΄ + π΅)π₯(π‘) (the general case of the classical price Rayleigh equation (2)) is exponentially stable (this is guaranteed by condition (16) and π min (π) > 0), then the corresponding delayed system (55) is also exponentially stable provided that the time delay π is small enough (bounded by (57)). Remark 12. Condition (17) controls the intensity of the uncertainties, that is, the parameters πΌπ , π½π (π = 1, 2, 3), and the time delay π should be small enough to have the stability of (13) which is regarded as the perturbed system of (55). In other words, Theorem 5 shows that if (55) is exponentially stable, so is its perturbed system (13) provided that the intensity of perturbations is small enough. Case 3. Let us assume that π΅ = π΄ 2 (π‘ β π) = π΅1 (π‘) = π΅2 (π‘ β π) β‘ 0 and π β‘ 0, then (13) reduces to a stochastic differential equation ππ₯ (π‘) = (π΄ + π΄ 1 (π‘)) π₯ (π‘) ππ‘ + β (π‘, π₯ (π‘)) ππ (π‘) .
(58)
Corollary 13. Assume that there exists a pair of symmetric positive definite π Γ π matrices πΊ and π such that πΊπ΄ + π΄π πΊ = βπ,
(59)
π min (π) > βπΊβ (2πΌ1 + πΌ3 ) .
(60)
Then, (58) is exponentially stable in the mean square and is also almost surely exponentially stable.
4. Example Let us now present two simple examples to illustrate our results which can help us find the time delay upper limit. Example 14. Let us start with (13), where πΌ1 = πΌ2 = πΌ3 = π½1 = π½2 = π½3 = 0.1, π΄=[
β1 0 ], β1 β1
π΅=[
β2 0 ]. 1 β1
βπ΄β2 = 2.618,
(61)
βπ΅β2 = 5.2361,
βπΊβ = 0.25,
βπΊπ΅β = 0.4488.
(63)
Substituting (63) into (17), we derive that if π < 0.08380, then (13) is exponentially stable in the mean square and is also almost surely exponentially stable. If β(π‘, π₯(π‘), π₯(π‘ β π)) β‘ 0, by Corollary 9, we would conclude that (52) is exponentially stable provided that π < 0.09156. Example 15. Now, let us recall price system (4) with π = β0.5, π0 = 1.0, π0 = 1.0, πΎ = 0.2, π = 0.1, βππΌ = 1, and πΜ = 0.06. It is clear that π§(π‘) = (0, 0)π is the trivial solution of system (4). Applying Taylor expansion to π(π‘, π§(π‘)), we get π (π‘, π§ (π‘)) = π (π‘, 0) + π΄π§ (π‘) + π΄ 1 (π‘) π§ (π‘) ,
(64)
π 1 where π΄ = [ 0.006 β1 0 ], tr[β (π‘, π₯(π‘))β(π‘, π₯(π‘))] = 0.004, and βπ΄ 1 (π‘)β β€ 0.001.
To compute conveniently, let us choose πΌ3 = 0.004 and 0 π = [ 0.01 0 0.01 ], and then π min (π) = 0.01. By Corollary 13, it is easy to find πΊ=[
1.66667 β0.00500 ], β0.00500 1.66670
βπΊβ = 0.5285.
(65)
Substituting (65) into (60), it is easy to verify that (60) holds. So, the price system (4) with π = β0.5, π0 = 1.0, π0 = 1.0, πΎ = 0.2, π = 0.1, βππΌ = 1, and πΜ = 0.06 is exponentially stable in the mean square and is also almost surely exponentially stable. This result is the same as [5].
5. Concluding Remarks In this paper, we study the exponential stability of the stochastic nonlinear dynamical price system. Using Taylorβs theorem, the stochastic nonlinear system with delay is reduced to an ndimensional semilinear stochastic differential equation with delay. Some sufficient conditions of exponential stability and corollaries for the price system are established by virtue of Lyapunov function. The time delay upper limit is solved by using our theoretical results when the system is exponentially stable, and [5] is promoted and improved. Remarks 11 and 12 show that if price Rayleigh equation (2) is exponentially stable, so is its perturbed system (13) provided that both the time delay and the intensity of perturbations are small enough. These results are very helpful for our government strengthening and improving macrocontrol and promoting
Mathematical Problems in Engineering steady and rapid economic development. It is also an important guiding significance that our government can timely adjust their pricing strategies. Two examples are presented to illustrate our theoretical results, which are the same as [5]. Another challenging problem is to study a type of stochastic nonlinear dynamical price system with variable delay. We hope to study these problems in forthcoming papers.
Acknowledgments This work is Supported by the Fundamental Research Funds for the Central Universities (JBK130213) and the Fundamental Research Funds for the Central Universities (JBK130401).
References [1] S. Wang, Differential Equation Model and Chaos, China Science and Technology University Press, Hefei, China, 2002. [2] Y. Liu and Z. Feng, βStochastic stability & control,β in Large Power System Theory and Application, South China University of Science and Technology Press, Guangzhou, China, 4th edition, 1992. [3] Y. Liu and F. Deng, βRandom system variable structure system,β in Large Power System Theory and Application, South China University of Science and Technology Press, Guangzhou, China, 10th edition, 1992. [4] J. Yong and X. Y. Zhou, Stochastic Controls: Hamiltonian Systems and HJB Equations, vol. 43 of Applications of Mathematics, Springer, New York, NY, USA, 1999. [5] J. R. Li and W. Xu, βOptimal control of a stochastic nonlinear dynamical price model,β Pure and Applied Mathematics, vol. 24, no. 2, pp. 239β244, 2008. [6] Y. Kazmerchuk, A. Swishchuk, and J. Wu, βThe pricing of options for securities markets with delayed response,β Mathematics and Computers in Simulation, vol. 75, no. 3-4, pp. 69β79, 2007. [7] T. Lv and Z. W. Liu, βHopf bifurcation of price Rayleigh equations with delay,β Journal of Jilin University, vol. 47, no. 3, pp. 441β448, 2009. [8] T. Lv and L. Zhou, βHopf and codimension two bifurcation in price Rayleigh equation with two time delay,β Journal of Jilin University, vol. 50, no. 3, pp. 441β448, 2012. [9] X. Zhang, X. Chen, and Y. Chen, βA qualitative analysis of price model in differential equations of price,β Journal of Shenyang Institute of Aeronautical Engineering, vol. 21, no. 1, pp. 83β86, 2004. [10] X. Mao, Exponential Stability of Stochastic Differential Equations, vol. 182 of Monographs and Textbooks in Pure and Applied Mathematics, Marcel Dekker, New York, NY, USA, 1994. [11] W. Zhu, J. Hu, and J. Zhang, βStability analysis of neural networks with time delays via energy functions approach,β in Proceedings of the 7th International Conference on Natural Computation (ICNC β11), vol. 1, pp. 232β236, 2011. [12] W. Zhu and Z. Yi, βIntegral input-to-state stability of nonlinear control systems with delays,β Chaos, Solitons & Fractals, vol. 34, no. 2, pp. 420β427, 2007. [13] H. Trinh and M. Aldeen, βOn robustness and stabilization of linear systems with delayed nonlinear perturbations,β IEEE Transactions on Automatic Control, vol. 42, no. 7, pp. 1005β1007, 1997.
9 [14] X. Mao, βRobustness of exponential stability of stochastic differential delay equations,β IEEE Transactions on Automatic Control, vol. 41, no. 3, pp. 442β447, 1996. [15] X. Mao and A. Shah, βExponential stability of stochastic differential delay equations,β Stochastics and Stochastics Reports, vol. 60, no. 1-2, pp. 135β153, 1997. [16] W. Zhu and J. Hu, βStability analysis of stochastic delayed cellular neural networks by LMI approach,β Chaos, Solitons & Fractals, vol. 29, no. 1, pp. 171β174, 2006. [17] W. Zhu and J. Hu, βExponential stability of stochastic recurrent neural network with time delays,β International Journal of Computational Intelligence Research, vol. 2, no. 1, pp. 52β54, 2006. [18] S. Xie and L. Xie, βStabilization of a class of uncertain large-scale stochastic systems with time delays,β Automatica, vol. 36, no. 1, pp. 161β167, 2000. [19] B. Γksendal, Stochastic Differential Equations: An Introduction with Applications, Springer, Berlin, Germany, 5th edition, 1998. [20] W. Zhu, J. Hu, P. Li, and C. Zhu, βOn the energy functions of neural networks with delay,β Dynamics of Continuous, Discrete & Impulsive Systems B, vol. 18, no. 5, pp. 589β599, 2011.
Advances in
Operations Research Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Advances in
Decision Sciences Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Journal of
Applied Mathematics
Algebra
Hindawi Publishing Corporation http://www.hindawi.com
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Journal of
Probability and Statistics Volume 2014
The Scientific World Journal Hindawi Publishing Corporation http://www.hindawi.com
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
International Journal of
Differential Equations Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Volume 2014
Submit your manuscripts at http://www.hindawi.com International Journal of
Advances in
Combinatorics Hindawi Publishing Corporation http://www.hindawi.com
Mathematical Physics Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Journal of
Complex Analysis Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
International Journal of Mathematics and Mathematical Sciences
Mathematical Problems in Engineering
Journal of
Mathematics Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Discrete Mathematics
Journal of
Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Discrete Dynamics in Nature and Society
Journal of
Function Spaces Hindawi Publishing Corporation http://www.hindawi.com
Abstract and Applied Analysis
Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
International Journal of
Journal of
Stochastic Analysis
Optimization
Hindawi Publishing Corporation http://www.hindawi.com
Hindawi Publishing Corporation http://www.hindawi.com
Volume 2014
Volume 2014