Weak Convergence of Markov Random Evolutions in a Multidimensional Space
Igor V. Samoilenko1
Academic Editor: A. Pascucci, E. Orsingher
Received04 May 2012
Accepted05 Jun 2012
Published16 Aug 2012
Abstract
We study Markov symmetrical and nonsymmetrical random evolutions in . Weak convergence of Markov symmetrical random evolution to Wiener process and of Markov non-symmetrical random evolution to a diffusion process with drift is proved using problems of singular perturbation for the generators of evolutions. Relative compactness in of the families of Markov random evolutions is also shown.
1. Introduction
Markov symmetrical random evolutions (MSRE) in spaces of different dimensions were studied in the works of Kac [1], Pinsky [2], Orsingher (e.g., [3, 4]), and Kolesnik and Turbin (e.g., [5, 6]) (see also [7] for other references). Symmetry in this sense should be regarded as uniform stationary distribution of switching at a symmetrical structure in , for instance at -hedron [7], or at a unit sphere [8].
Weak convergence of distributions of MSRE is also studied in some of these works, namely, convergence in and was proved by Kolesnik in [8, 9].
In this work we not only generalize the results of Kolesnik on the multidimensional space, but solve another problem. We prove weak convergence of the process of Markov random evolution, that means not only proof of weak convergence of respective distributions (or generators) but also proof of compactness of the process. In majority of works that deal with random evolutions, compactness of the process is not considered at all. It is also shown that the symmetry of the process is closely connected with balance condition (see (2.23), (4.4)).
We should note that the problem of weak convergence of random walks (partially, similar to MSRE) was studied by many authors. Among the most interesting works, we can point out [10β13]. Large bibliography as for this problem could be found in [10]. The methods proposed in these works allow us to solve a wide range of problems connected with convergence of random walks but do not let us obtain limit process to be averaged by the stationary measure of switching process.
Such averaging can be found in the works of Anisimov and his students (see [14] and references therein), but here the averaging by the stationary measure is one of the conditions proposed for the prelimit process in the corresponding theorem.
In this work, MSRE in is studied using the methods proposed in [15]. We find a solution of singular perturbation problem for the generator of the evolution, and thus the averaging by a stationary measure of switching process is obtained as a corollary of this solution. At the second stage we prove relative compactness of the family of MSRE. This method let us show weak convergence of the process of MSRE to the Wiener process in .
The difference in the methods can be easily seen by the analysis of the papers [16, 17], where similar problems are studied.
In Sections 3 and 4, we use the method proposed to prove weak convergence of Markov non-symmetrical random evolution (MNRE) in . The distinction of this model is that the limit process is a diffusion process with drift.
2. Description of MSRE
We study a particle in the space , that starts at from the point . Possible directions of motion are given by the following vectors:
These vectors have initial point in the center of the unit -dimentional sphere and the terminal point at its surface. Choosing of every next direction is random and its time is distributed by the Poisson law. Thus, the switching process process is the Poisson one with intensity . The velocity of particleβs motion is fixed and equals , where is a small parameter, .
Let us define a set and suppose be the switching Poisson process.
Definition 2.1. Markov symmetrical random evolution (MSRE) is the process , given by
It is easy to see that when the velocity of the particle and intensity of switching decrease. Our aim is to prove weak convergence of MSRE to the Wiener process when . The main method is solution of singular perturbation problem for the generator of MSRE. Let us describe this generator. Two-component Markov process at the test functions can be described by the generator (see, e.g., [2])
where
here ,
and for even , for odd ; is the element of volume of the sphere , that is equal to
Using well-known formula
we can see
Operator is the projector at the null space of reducibly invertible operator , because by definition it maps functions to constants but constants to itself. For we have
Potential operator can be defined as the following:
This operator has the following property:
thus it is inverse for at the range of , but for the function from the null space of we have
Solution of singular perturbation problem in the series scheme with the small series parameter (see [15]) for reducibly invertible operator and perturbing operator consists in the following. We should find two vectors and , that satisfy asymptotic representation
with the vector , that is uniformly bounded by the norm and such that
The left part of the equation can be rewritten as
And as soon as it is equal to the right side, we obtain
From the last equation we may see that the function should be smooth enough to provide boundness of . Moreover, from the first equation we see that any function from the null space of can be taken as , and it does not depend on the variable that corresponds to the switching process. An important condition of solvability for this problem is the balance condition
This condition means that the function belongs to the range of , thus we may solve the second equation using the potential operator, that is, inverse to at its range
Thus, the main problem is to solve the following equation:
The solvability condition for has the view
and we finally obtain
For the function we obviously have
Equations (2.18)β(2.22) give the solution of singular perturbation problem. In case of MSRE, the balance condition has the following view:
where . Really, every term under the integral contains either or .
3. Main Result for MSRE
Theorem 3.1. MSRE , converges weakly to the Wiener process when as
where is defined by the generator
Here .
Remark 3.2. The generator (3.2) generalizes the generators obtained in [8] for the spaces and . We use the following Lemma to prove the Theorem.
Lemma 3.3. At the perturbed test functions
having bounded derivatives of any degree and compact support, the operator has the asymptotic representation
where is defined in (3.2), and are the following:
Proof. Let us solve singular perturbation problem for the operator (2.3). To do this, we use the following correlation:
Thus, we obtain equations
From the first equation we see that belongs to the null space of . It is easy to see from the balance condition (2.23) that belongs to the range of , thus from the second equation of the system (3.7) we have
By substitution into the third equation and using the solvability condition, we can see
From the last equation of (2.23) we have
Letβs find the generator of the limit process by the formula (3.5) as
The last term equals to 0 by the balance condition (2.23). Thus, finally
or
Let us calculate the following integral:
Every term in braces has a multiplier of the type or , thus the corresponding integral equals to 0. Integration of the correlations in the parentheses gives
Finally, we have
Lemma is proved.
Proof of Theorem 3.1. In Lemma 3.3, we proved that at the class of functions when . To prove the weak convergence of the process, we should show relative compactness of the family in . To do this, we use the methods proposed in [15, 18, 19]. Let us formulate Corollary 6.1 from [15] (see also Theoremββ6.4 in [15]) as Lemma 3.4.
Lemma 3.4. Let the generators , satisfy the inequalities
for any real-valued nonnegative function , where the constant depends on the norm of , and for ,
where the constant depends on the function , but does not depend on . Then, the family , , is relatively compact in .
Let us write the action of the generator (2.3) at the test function , where .
We have
It follows from (3.7) that the first two terms equal to 0. Let us estimate the following last term:
as soon as all the constants, functions, and their derivatives are bounded.
To prove the second condition, itβs enough to use the properties of the function , namely:
So, the proof of the second condition is similar to the previous reasoning.
Thus, the family is relatively compact in .
Now we may use the following theorem (Theoremββ6.6 from [15]).
Theorem 3.5. ββLet random evolution with Markov switching satisfies the following conditions.(C1)The family of processes , , is relatively compact.(C2)There exists the family of test-functions , such that
βuniformly by , .(C3) The following uniform convergence is true
βuniformly by .βThe family , is uniformly bounded, moreover and belong to .(C4)Convergence by probability of initial values
βis true. Then we have the weak convergence
According to Theorem 3.5, we may confirm the weak convergence in
Really, all the conditions are satisfied. Namely, the family of processes is relatively compact, the generators converge at the test functions belonging to the class , initial conditions for the limit, and prelimit processes are equal.
Theorem is proved.
4. Description of MNRE
We study the same particle in , but its velocity is equal to , where is the small parameter, the functions are bounded.
Definition 4.1. Markov nonsymmetrical random evolution (MNRE) is the process , given by
Our aim is to prove the weak convergence of MNRE to a diffusion process with drift when . Two-component Markov process at the test functions is defined by a generator (see, e.g., [2])
where
An important condition that let us confirm weak convergence is balance condition
Remark 4.2. This is the last condition that defines symmetry or nonsymmetry of the process. In case of MSRE, the balance condition (4.4) is true, and . In case of MNRE the absence of symmetry of the process is caused by the following condition:
Example 4.3. Condition (4.4) can be satisfied for different functions . Namely, in case of MSRE . Then every term under the integral contains either or . Another example is the function . Really, we obtain the terms under the integral that are analogical to the previous ones. We note that the dimension of the space in this case should be more than 2, because in this function does not have symmetry.
Example 4.4. Condition (4.5) can be also satisfied for different functions . For example, in for we obtain
Another example is the following function in :
Again, all terms under the integral, except the last one, contain , thus only one term is not trivial as
By simple calculations, we see
We obviously have a wide range of functions that preserve or, on the contrary, do not preserve symmetry. So, we can define the velocity of random evolution in different ways, depending on possible applications.
5. Main Result for MNRE
Theorem 5.1. MNRE , converges weakly to the process when as
The limit process is defined by a generator
where ,
We need the following Lemma to prove the Theorem.
Lemma 5.2. At the perturbed test functions
having bounded derivatives of any degree and compact support, the operator has the following asymptotic representation:
where is defined in (5.2), , and are defined by the following correlations:
Proof. We solve singular perturbation problem for the operator (4.2). Using the view of test-functions (5.4), we have
Thus, we obtain the following equations:
According to the first one, belongs to the null space of . From the balance condition (4.4) we see that belongs to the range of , thus from the second equation of (5.8) we have
By substitution into the third equation and using the solvability condition, we obtain:
From the last equation, we have
We calculate the operator by the formula (5.6) as
The second term equals 0 by the balance condition (4.4), the last one is not equal to 0 by (4.5). Thus, we finally have
Using the view of , we may write
Lemma is proved.
Proof of Theorem 5.1. We proved in Lemma 5.2 that at the class of test functions , when . To prove the weak convergence, we should show relative compactness of the family of processes in . To do this, we use Lemma 3.4. We have the following view of the operator (4.2) at the test function , where :
It follows from (5.8) that the first two terms equal to 0. Let us estimate the following third term:
as soon as all the constants, functions, and their derivatives are bounded. The last term may be estimated analogically. From (5.8) we also have
Thus,
for that is small enough. To prove the second condition of Lemma 3.4, we use the following properties of the function :
We can see again that the proof of the second condition is similar to the previous reasoning. Thus, the family of the processes is relatively compact in . Using Theorem 3.5, we confirm the following weak convergence in :
Really, all the conditions are satisfied. Namely, the family of processes is relatively compact, the generators at the test functions belonging to the class converge, and initial conditions for the limit and prelimit processes are equal. Theorem is proved.
Example 5.3. Let us consider one more variant of evolution in . Let
In other cases both functions equal to 0. The balance condition (4.4) is true for , on the contrary, condition (4.5) is true for . The limit generator (5.2) has the view
Thus, the limit process has two parts: the drift with velocity in direction of coordinate and diffusion part in one-dimensional subspace, corresponding to coordinate that is similar to the limit process described in Kac model [1].
List of Symbols
:
the components of velocity of random evolutionβs motion
:
the space of bounded continuous functions on having continuous derivatives of all orders vanishing at infinity
:
constant depending on the norm of function
:
parameter of drift
:
the space of -valued right continuous functions having left limits (cadlag), defined on
:
the generator of random evolution
:
the generator of limit process
:
volume of
:
dimension of space
:
the generator of switching Poisson process
:
perturbing operator
:
potential of generator
:
negligible term
:
unit -dimensional sphere
:
operator that describes particleβs motion
:
vectors that define possible directions of random evolutionβs motion
:
velocity of random evolutionβs motion
:
small parameter
,ββ,ββ,ββ:
angles that define directions in
:
the set of all angles
:
Poisson process that switches directions of random evolutionβs motion
:
intensity of switching Poisson process
:
small element of volume of
:
Markov random evolutions
:
limit processes
:
projector at the nullspace of generator
:
parameter of diffusion.
References
M. Kac, βA stochastic model related to the telegrapher's equation,β The Rocky Mountain Journal of Mathematics, vol. 4, pp. 497β509, 1974.
M. A. Pinsky, Lectures on Random Evolution, World Scientific Publishing, Singapore, 1991.
E. Orsingher, βBessel functions of third order and the distribution of cyclic planar motions with three directions,β Stochastics and Stochastics Reports, vol. 74, no. 3-4, pp. 617β631, 2002.
E. Orsingher and A. M. Sommella, βA cyclic random motion in with four directions and finite velocity,β Stochastics and Stochastics Reports, vol. 76, no. 2, pp. 113β133, 2004.
A. D. Kolesnik and A. F. Turbin, βInfinitesimal hyperbolic operator of Markovian random evolutions,β Doklady Akademii Nauk Ukrainskoj, vol. 1, pp. 11β14, 1991 (Russian).
A. D. Kolesnik, βWeak convergence of the distributions of Markovian random evolutions in two and three dimensions,β Buletinul Academiei de Ştiinţe a Republicii Moldova. Matematica, no. 3, pp. 41β52, 2003.
A. D. Kolesnik, βWeak convergence of a planar random evolution to the Wiener process,β Journal of Theoretical Probability, vol. 14, no. 2, pp. 485β494, 2001.
A. A. Borovkov and K. A. Borovkov, Asymptotic Analysis of Random Walks, vol. 118 of Encyclopedia of Mathematics and its Applications, Cambridge University Press, Cambridge, UK, 2008.
K. Borovkov, βOn random walks with jumps scaled by cumulative sums of random variables,β Statistics & Probability Letters, vol. 35, no. 4, pp. 409β416, 1997.
S. Foss, T. Konstantopoulos, and S. Zachary, βDiscrete and continuous time modulated random walks with heavy-tailed increments,β Journal of Theoretical Probability, vol. 20, no. 3, pp. 581β612, 2007.
S. Zakhari and S. G. Foss, βOn the exact asymptotics of the maximum of a random walk with increments in a class of light-tailed distributions,β Siberian Mathematical Journal, vol. 47, no. 6, pp. 1034β1041, 2006.
V. V. Anisimov, Switching Processes in Queueing Models, Applied Stochastic Methods Series, ISTE, London, UK, John Wiley & Sons, Hoboken, NJ, USA, 2008.
V. V. Anīsīmov, βConvergence of accumulation processes with switchings,β Teorīya Imovīrnosteĭ ta Matematichna Statistika, no. 63, pp. 3β12, 2000, translation in Theory of Probability and Mathematical Statistics no. 63, 1–11,
2001-2002.
I. V. Samoilenko, βConvergence of an impulsive storage process with jump switchings,β Ukraïni Matematicni Zhurnal, vol. 60, no. 9, pp. 1282β1286, 2008, translation in Ukrainian Mathematical Journal, vol. 60, no. 9,
1492–1497, 2008.
S. N. Ethier and T. G. Kurtz, Markov Processes: Characterization and Convergence, Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics, John Wiley & Sons, New York, NY, USA, 1986.
D. W. Stroock and S. R. S. Varadhan, Multidimensional Diffusion Processes, vol. 233 of Fundamental Principles of Mathematical Sciences, Springer, Berlin, Germany, 1979.