- Research
- Open access
- Published:
On the p th moment estimates for the solution of stochastic differential equations
Journal of Inequalities and Applications volume 2014, Article number: 395 (2014)
Abstract
The main aim of this paper is to discuss the -estimate of solution of stochastic differential equations under nonlinear growth condition. Furthermore, almost surely asymptotic estimates for the solutions of these equations are given. More accurately, we shall estimate sample Lyapunov exponent almost surely.
MSC:60H10, 60H30.
1 Introduction
Stochastic analysis including Brownian motion processes has played an important role in many areas of science and engineering for a long time. Since the white noise is mathematically represented by a formal derivative of a Brownian motion process, such stochastic analysis is based on various types of stochastic differential equations(SDEs) of Itô type as a stochastic model. After Itô introduced his stochastic calculus, the theory of SDEs has been developed very quickly. SDEs is the most fundamental concept in modern stochastic models.
Consequently, there is an increasing interest in SDEs. The main interest in the field has often referred to the existence and uniqueness of solutions as well as to study of their qualitative and quantitative properties, with a special emphasis on the analysis of various types of -estimate, almost sure estimate, stability and approximation. We refer the reader to monographs [1] by Henderson and Plaschko, [2] by Kolmanovskii and Myshkis, and [3] by Mao, among others, and the literature cited therein.
It is well known that the problems of the solutions to SDEs have received considerable attention from the theoretical point of view. One of the classical results and a subject for inquiry in the study of SDEs is an existence and uniqueness theorem of the solution to SDEs under some special conditions. For the work on the existence and uniqueness theorem of the SDEs, we refer the reader to [4] by Cho et al., [5] by Ren and Xia, [6] by Wei and Wang, and [3] by Mao. Now there is an extensive literature discussing SDEs with p th moment estimate and stability of the solution (see [7] by Govindan, [8] by Kim, [9] by Li and Fu, and [10] by Mao et al.).
Further, in the study of the solution for the SDEs, one question arises naturally: Does the p th moment of the solution assure the solution for such SDEs? To the best of our knowledge, there are few results on this problem. It is also worth noting that the p th moment of the solution for such SDEs has not been fully investigated, which remains an interesting research topic.
Our study is essentially based on [3] by Mao referring to the -estimate and almost surely asymptotic estimates of SDEs. In fact, we generalize Mao’s results by finding sufficient conditions, which are easy to verify, guaranteeing -estimate and almost surely asymptotic estimates of the solutions of these equations.
2 Preliminary
Let denote an Euclidean norm in . If A is a vector or a matrix, its transpose is denoted by ; if A is a matrix, its trace norm is represented by . Let be a positive constant and throughout this paper, unless otherwise specified, be a complete probability space with a filtration satisfying the usual conditions (i.e., it is right continuous and contains all P-null sets). Assume that is an m-dimensional Brownian motion defined on a complete probability space, that is, . Let denote the family of -valued -adapted processes such that a.s.
Consider the d-dimensional stochastic differential equations of Itô type
with initial value , , and be Borel measurable.
In [3], the author presented a result stating that, for initial value , the p th moment of the unique solution , of equation (2.1) under Lipschitz condition and linear growth condition will grow at most exponentially with exponent pα. We reproduce the result here.
Theorem 2.1 Let and . Assume that there exists a constant α such that for all ,
Then
3 Main results
The topic of our analysis is equation (2.1) with initial data . An -adapted process with values in is said to be the solution to equation (2.1) if it satisfies the initial condition and the corresponding stochastic integral equation holds a.s., i.e., for every ,
The basic existence and uniqueness theorem based on the Picard method of iterations requires the global or, in a weakened version, Lipschitz condition and linear growth condition for f and g (see [3–6]). Then there exists a unique a.s. continuous and adapted solution to equation (2.1) satisfying for every under the linear growth condition. Moreover, if , , then under the monotone condition. Since our goal is to study -estimate and almost surely asymptotic estimate problems, we assume that there exists a unique solution to equation (2.1) under non-Lipschitz condition and nonlinear growth condition (see [5, 6]). We also assume that all the Lebesgue and Itô’s integrals employed further are well defined.
We start with the following -estimate.
Theorem 3.1 Let and . Assume that for all , , and , it follows that
where is a concave nondecreasing function from to such that , for . Then, for a pair of positive constants a and b such that , we have
for all .
Proof By Itô’s formula, we can derive that for ,
By condition (3.1), it is easy to see that
Given that is concave and , we can find a pair of positive constants a and b such that for all . And for each number , define the stopping time
Obviously, as , a.s. Moreover, it follows from (3.3) and the property of Itô’s integral that
The Gronwall inequality yields
Letting yields
and the desired inequality follows. □
Let us now turn to considering the case of . This is rather easy if we note that the Hölder inequality implies
In other words, the estimate for can be done via the estimate for the second moment. For instance, we have the following corollary.
Corollary 3.2 Let and . Assume that for all , , and , it follows that
where is a concave nondecreasing function from to such that , for . Then, for a pair of positive constants a and b such that , we have
for all .
We now verify that if a nonlinear growth condition
with is a concave nondecreasing function from to such that , for is fulfilled, then condition (3.1) is satisfied with . In fact, using (3.7) and the elementary inequality , one can derive that for any ,
Letting yields
Consequently,
We therefore obtain the following useful corollary.
Corollary 3.3 Let and . Assume that the nonlinear growth condition (3.7) holds. Then inequality (3.1) holds with
We now apply this result to show one of the important properties of the solution.
Theorem 3.4 Let and . Assume that the nonlinear growth condition (3.7) holds for all . Then
where . In particular, the pth moment of the solution is continuous on .
Proof Applying the elementary inequality , we can easily see that
Using the Hölder inequality, the moment inequality ([3], Theorem 1.7.1), and condition (3.7), one can show that
where . Given that is concave and , we can find a pair of positive constants α and β such that for all . So we have
Substituting this into (3.4) yields that
which is the required inequality. □
Theorem 3.5 Let and . Assume that the nonlinear growth condition (3.7) for all . Then
for all , where and .
Proof Using the Hölder inequality, the moment inequality ([3], Theorem 1.7.2), and condition (3.7), one can show that
where . From the definition of , we can find a pair of positive constants α and β such that for all . So we have
Hence
By the Gronwall inequality one can see that
which is the required inequality. □
Let us now consider a d-dimensional stochastic differential equation
on with initial value . Assume that the equation has a unique global solution on . Moreover, we shall impose the condition: There is a concave nondecreasing function from to such that, for all ,
Let . In view of Theorem 3.1 and Corollary 3.2, we know that the p th moment of the solution satisfies
for all . This means that the p th moment will grow at most exponentially with exponent . This can also be expressed as
The left-hand side of (3.10) is called the pth moment Lyapunov exponent, and (3.10) shows that the p th moment Lyapunov exponent should not be greater than . In what follows, we shall establish the asymptotic estimate for the solution almost surely. More accurately, we shall estimate
almost surely, which is called the sample Lyapunov exponent.
Theorem 3.6 Under condition (3.9), the sample Lyapunov exponent of the solution of equation (3.8) should not be greater than , that is,
almost surely.
Proof By Itô’s formula and condition (3.9), we obtain
where
Furthermore, for every integer , using the exponential martingale inequality (see [3], Theorem 1.7.4) on sees that
An application of the Borel-Cantelli lemma then yields that for almost all , there is a random integer such that
if . That is,
for all , almost surely. From inequality (3.11) and the definition of the function , we deduce that
for all , almost surely. Therefore, for almost all , if , ,
This implies
almost surely, which is the required inequality. □
References
Henderson D, Plaschko P: Stochastic Differential Equations in Science and Engineering. World Scientific, Singapore; 2006.
Kolmanovskii VB, Myshkis A: Applied Theory of Functional Differential Equations. Kluwer Academic, Dordrecht; 1992.
Mao X: Stochastic Differential Equations and Applications. Horwood, Chichester; 2007.
Cho YJ, Dragomir SS, Kim Y-H: A note on the existence and uniqueness of the solutions to SFDEs. J. Inequal. Appl. 2012. Article ID 126, 2012: Article ID 126
Ren Y, Xia N: Existence, uniqueness and stability of the solutions to neutral stochastic functional differential equations with infinite delay. Appl. Math. Comput. 2009, 210: 72-79. 10.1016/j.amc.2008.11.009
Wei F, Wang K: The existence and uniqueness of the solution for stochastic functional differential equations with infinite delay. J. Math. Anal. Appl. 2007, 331: 516-531. 10.1016/j.jmaa.2006.09.020
Govindan TE: Stability of mild solution of stochastic evolution equations with variable delay. Stoch. Anal. Appl. 2003, 21: 1059-1077. 10.1081/SAP-120022863
Kim Y-H: An estimate of the solutions for stochastic functional differential equations. J. Appl. Math. Inform. 2011,29(5-6):1549-1556.
Li X, Fu X: Stability analysis of stochastic functional differential equations with infinite delay and its application to recurrent neural networks. J. Comput. Appl. Math. 2010, 234: 407-417. 10.1016/j.cam.2009.12.033
Mao X, Shen Y, Yuan C: Almost surely asymptotic stability of neutral stochastic differential delay equations with Markovian switching. Stoch. Process. Appl. 2008, 118: 1385-1406. 10.1016/j.spa.2007.09.005
Acknowledgements
This research is financially supported by Changwon National University in 2013-2014.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The author declares that they have no competing interests.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Kim, YH. On the p th moment estimates for the solution of stochastic differential equations. J Inequal Appl 2014, 395 (2014). https://doi.org/10.1186/1029-242X-2014-395
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/1029-242X-2014-395