Processing math: 92%

Your new experience awaits. Try the new design now and help us make it even better

ORIGINAL RESEARCH article

Front. Phys., 14 January 2022

Sec. Interdisciplinary Physics

Volume 9 - 2021 | https://doi.org/10.3389/fphy.2021.795210

This article is part of the Research TopicLong-Range Dependent Processes: Theory and ApplicationsView all 15 articles

Large Time Behavior on the Linear Self-Interacting Diffusion Driven by Sub-Fractional Brownian Motion With Hurst Index Large Than 0.5 I: Self-Repelling Case

Han Gao
Han Gao1*Rui GuoRui Guo2Yang JinYang Jin3Litan YanLitan Yan3
  • 1College of Fashion and Art Design, Donghua University, Shanghai, China
  • 2College of Information Science and Technology, Donghua University, Shanghai, China
  • 3Department of Statistics, College of Science, Donghua University, Shanghai, China

Let SH be a sub-fractional Brownian motion with index 12<H<1. In this paper, we consider the linear self-interacting diffusion driven by SH, which is the solution to the equation

dXHt=dSHtθ(t0(XHtXHs)ds)dt+νdt,XH0=0,

where θ < 0 and νR are two parameters. Such process XH is called self-repelling and it is an analogue of the linear self-attracting diffusion [Cranston and Le Jan, Math. Ann. 303 (1995), 87–93]. Our main aim is to study the large time behaviors. We show the solution XH diverges to infinity, as t tends to infinity, and obtain the speed at which the process XH diverges to infinity as t tends to infinity.

1 Introduction

In 1995, Cranston and Le Jan [1] introduced a linear self-attracting diffusion

Xt=Btθt0s0(XsXu)duds+νt,t0

with θ > 0 and X0 = 0, where B is a 1-dimensional standard Brownian motion. They showed that the process Xt converges in L2 and almost surely, as t tends infinity. This is a special case of path dependent stochastic differential equations. Such path dependent stochastic differential equation was first developed by Durrett and Rogers [2] introduced in 1992 as a model for the shape of a growing polymer (Brownian polymer) as follows

Xt=X0+Bt+t0s0f(XsXu)duds,

where B is a d-dimensional standard Brownian motion and f is Lipschitz continuous. Xt corresponds to the location of the end of the polymer at time t. Under some conditions, they established asymptotic behavior of the solution of stochastic differential equation and gave some conjectures and questions. The model is a continuous analogue of the notion of edge (resp. vertex) self-interacting random walk. If f(x) = g(x)x/‖x‖ and g(x) ≥ 0, Xt is a continuous analogue of a process introduced by Diaconis and studied by Pemantle [3]. Let LX(t,x) be the local time of the solution process X. Then, we have

Xt=X0+Bt+t0dsRf(x)LX(s,Xs+x)dx

for all t ≥ 0. This formulation makes it clear how the process X interacts with its own occupation density. We may call this solution a Brownian motion interacting with its own passed trajectory, i.e., a self-interacting motion. In general, the Eq. 1.2 defines a self-interacting diffusion without any assumption on f. If

xf(x)0(xf(x)0)

for all xRd, we call it self-repelling (resp. self-attracting). In 2002, Benaïm et al [4] also introduced a self-interacting diffusion with dependence on the (convolved) empirical measure. A great difference between these diffusions and Brownian polymers is that the drift term is divided by t. It is noteworthy that the interaction potential is attractive enough to compare the diffusion (a bit modified) to an Ornstein-Uhlenbeck process, in many case of f, which points out an access to its asymptotic behavior. More works can be found in Benaïm et al. [5], Cranston and Mountford [6], Gauthier [7], Herrmann and Roynette [8], Herrmann and Scheutzow [9], Mountford and Tarr [10], Shen et al [11], Sun and Yan [12] and the references therein.

On the other hand, starting from the application of fractional Brownian motion in polymer modeling, Yan et al [13] considered an analogue of the linear self-interacting diffusion:

XHt=BHtθt0s0(XHsXHu)duds+νt,t0

with θ ≠ 0 and XH0=0, where BH is a fractional Brownian motion (fBm, in short) with Hurst parameter 12H<1. The solution of (1.3) is a Gaussian process. When θ > 0, Yan et al [13] showed that the solution XH of (1.3) converges in L2 and almost surely, to the random variable

XH=0hθ(s)dBHs+ν0hθ(s)ds

where the function is defined ar follows

hθ(s)=1θse12θs2se12θu2du,s0

with θ > 0. Recently, Sun and Yan [14] considered the related parameter estimations with θ > 0 and 12H<1, and Gan and Yan [15] considered the parameter estimations with θ < 0 and 12H<1.

Motivated by these results, as a natural extension one can consider the following stochastic differential equation:

Xt=Gtθt0s0(XsXu)duds+νt,t0

with θ > 0 and X0 = 0, where G = {Gt, t ≥ 0} is a Gaussian process with some suitable conditions which includes fractional Brownian motion and some related processes. However, for a (general) abstract Gaussian process it is difficult to find some interesting fine estimates associated with the calculations. So, in this paper we consider the linear self-attracting diffusion driven by a sub-fractional Brownian motion (sub-fBm, in short). We choose this kind of Gaussian process because it is only the generalization of Brownian motion rather than the generalization of fractional Brownian motion. It only has some similar properties of fractional Brownian motion, such as long memory and self similarity, but it has no stationary increment. The so-called sub-fBm with index H ∈ (0, 1) is a mean zero Gaussian process SH={SHt,t0} with SH0=0 and the covariance

RH(t,s)E[SHtSHs]=s2H+t2H12[(s+t)2H+|ts|2H]

for all s, t ≥ 0. For H = 1/2, SH coincides with the standard Brownian motion B. SH is neither a semimartingale nor a Markov process unless H = 1/2, so many of the powerful techniques from stochastic analysis are not available when dealing with SH. As a Gaussian process, it is possible to construct a stochastic calculus of variations with respect to SH (see, for example, Alós et al [16]). The sub-fBm has properties analogous to those of fBm and satisfies the following estimates:

[(222H1)1](ts)2HE[(SHtSHs)2][(222H1)1](ts)2H.

More works for sub-fBm and related processes can be found in Bojdecki et al. [1720], Li [2124], Shen and Yan [25, 26], Sun and Yan [27], Tudor [2831], Ciprian A. Tudor [32] Yan et al [3335] and the references therein.

In this present paper, we consider the linear self-interacting diffusion

XHt=SHtθt0s0(XHsXHu)duds+νt,t0

with θ < 0 and XH0=0, where SH is a sub-fBm with Hurst parameter 12H<1. Our main aim is to show that the solution of (1.7) diverges to infinity and obtain the speed diverging to infinity, as t tends to infinity. The object of this paper is to expound and prove the following statements:

(I) For θ < 0 and 12<H<1, the random variable

ξH=0se12θs2dSHs

exists as an element in L2.

(II) For θ < 0 and 12<H<1, as t, we have

JH0(t;θ,ν)te12θt2XHtξHνθ

in L2 and almost surely.

(III) For θ < 0 and 12<H<1, define the processes JH(n,θ,ν)={JHt(n,θ,ν),t0},n1 by

JHn(t;θ,ν)θt2(JHn1(t;θ,ν)(2n3)(ξHνθ)),n=1,2,,

for all t ≥ 0, where (−1)!! = 1. We then have

JHn(t;θ,ν)(2n1)(ξHνθ)

holds in L2 and almost surely for every n ≥ 1, as t.

This paper is organized as follows. In Section 2 we present some preliminaries for sub-fBm and Malliavin calculus. In Section 3, we obtain some lemmas. In Section 4, we prove the main result. In Section 5 we give some numerical results.

FIGURE 1
www.frontiersin.org

FIGURE 1. A path of XH with θ = − 1 and H = 0.7.

FIGURE 2
www.frontiersin.org

FIGURE 2. A path of XH with θ = − 10 and H = 0.7.

FIGURE 3
www.frontiersin.org

FIGURE 3. A path of XH with θ = − 100 and H = 0.7.

FIGURE 4
www.frontiersin.org

FIGURE 4. A path of XH with θ = − 1 and H = 0.5.

FIGURE 5
www.frontiersin.org

FIGURE 5. A path of XH with θ = − 10 and H = 0.5.

FIGURE 6
www.frontiersin.org

FIGURE 6. A path of XH with θ = − 100 and H = 0.5.

TABLE 1
www.frontiersin.org

TABLE 1. The data of XHt with θ = − 1 and H = 0.7.

TABLE 2
www.frontiersin.org

TABLE 2. The data of XHt with θ = − 10 and H = 0.7.

TABLE 3
www.frontiersin.org

TABLE 3. The data of XHt with θ = − 100 and H = 0.7.

TABLE 4
www.frontiersin.org

TABLE 4. The data of XHt with θ = − 1 and H = 0.5.

TABLE 5
www.frontiersin.org

TABLE 5. The data of XHt with θ = − 10 and H = 0.5.

TABLE 6
www.frontiersin.org

TABLE 6. The data of XHt with θ = − 100 and H = 0.5.

2 Preliminaries

In this section, we briefly recall the definition and properties of stochastic integral with respect to sub-fBm. We refer to Alós et al [16], Nualart [36], and Tudor [31] for a complete description of stochastic calculus with respect to Gaussian processes. Throughout this paper we assume that SH={SHt,t0} denotes a sub-fBm defined on the probability space (Ω,F,P) with index H. As we pointed out before, the sub-fBm SH is a rather special class of self-similar Gaussian processes such that SH0=0, E[SHt]=0 and

RH(t,s)E[SHtSHs]=s2H+t2H12[(s+t)2H+|ts|2H]

for all s, t ≥ 0. For H = 1/2, SH coincides with the standard Brownian motion B. SH is neither a semimartingale nor a Markov process unless H = 1/2, so many of the powerful techniques from stochastic analysis are not available when dealing with SH. As a Gaussian process, it is possible to construct a stochastic calculus of variations with respect to SH. The sub-fBm appeared in Bojdecki et al [17] in a limit of occupation time fluctuations of a system of independent particles moving in Rd according a symmetric α-stable Lévy process, and it also appears in Bojdecki et al [18] in a high-density limit of occupation time fluctuations of the above mentioned particle system, where the initial Poisson configuration has finite intensity measure.

The estimate (1.6) and normality imply that the sub-fBm tSHt admits almost surely a bounded 1Hϑ-variation on any finite interval for any sufficiently small ϑ ∈ (0, H). That is, the paths of tSHt admits a bounded pH-variation on any finite interval with pH>1H. As an immediate result, one can define the Young integral of a process u = {ut, t ≥ 0} with respect to sub-fBm Ba,b

t0usdSHs

as the limit in probability of a Riemann sum. Clearly, the integral is well-defined and

utSHt=t0usdSHs+t0SHsdus

for all t ≥ 0, provided u is of bounded qH-variation on any finite interval with qH > 1 and 1pH+1qH>1 (see, for examples, Bertoin [37] and FöIllmer [38]).

Let H be the completion of the linear space E generated by the indicator functions 1[0,t], t ∈ [0, T] with respect to the inner product

1[0,s],1[0,t]H=RH(t,s)

for s, t ∈ [0, T]. When 12<H<1, we can show that

φ2H=T0T0φ(t)φ(s)2tsRH(t,s)dsdt=T0T0φ(t)φ(s)ψH(t,s)dsdt,φH,

where

ψH(t,s)=2tsRa,b(t,s)=H(2H1)(|ts|2H2|t+s|2H2)

for s, t ∈ [0, T]. Define the linear mapping EφSH(φ) by

1[0,t]SH(1[0,t])=T01[0,t](s)dSHsSHt

for all t ∈ [0, T] and it can be continuously extended to H and we call the mapping Φ is called the Wiener integral with respect to SH, denoted by

SH(φ)=T0φ(s)dSHs

and

φ2H=E(T0φ(s)dSHs)2

for any φH.

For simplicity, in this paper we assume that 12<H<1. Thus, if for every T > 0, the integral

T0φ(s)dSHs

exists in L2 and

00φ(t)φ(s)ψH(t,s)dsdt<,

we can define the integral

0φ(s)dSHs

and

E(0φ(s)dSHs)2=00φ(t)φ(s)ψH(t,s)dsdt.

Denote by S the set of smooth functionals of the form

F=f(SH(φ1),SH(φ2),,SH(φn)),

where fCb(Rn) and φiH. The Malliavin derivative D of a functional F as above is given by

DF=nj=1fxj(SH(φ1),SH(φ2),,SH(φn))φj.

The derivative operator D is then a closable operator from L2(Ω) into L2(Ω;H). We denote by D1,2 the closure of S with respect to the norm

F1,2E|F|2+EDF2H.

The divergence integral δ is the adjoint of derivative operator DH. That is, we say that a random variable u in L2(Ω;H) belongs to the domain of the divergence operator δ, denoted by Dom(δS), if

E|DF,uH|cFL2(Ω)

for every FD1,2, where c is a constant depending only on u. In this case δ(u) is defined by the duality relationship

E[Fδ(u)]=EDF,uH

for any FD1,2. We have D1,2Dom(δ) and for any uD1,2

E[δ(u)2]=Eu2H+EDu,(Du)HH=Eu2H+E[0,T]4DξurDηusϕH(η,r)ϕH(ξ,s)dsdrdξdη,

where (DSu) is the adjoint of Du in the Hilbert space HH. We will denote

δ(u)=T0usδSHs

for an adapted process u, and it is called Skorohod integral. Alós et al [16], we can obtain the relationship between the Skorohod and Young integral as follows

T0usdSHs=T0usδSHs+T0T0Dsutψ(t,s)dsdt,

provided u has a bounded q-variation with 1q<1H and uD1,2(H) such that

T0T0Dsutψ(t,s)dsdt<.

Theorem 2.1 (Alós et al [16]). Let 0 < H < 1 and let fC2(R) such that

max{|f(x)|,|f(x)|,|f(x)|}κeβx2,

where κ and β are two positive constants with β<14T2H. Then we have

f(SHt)=f(0)+t0f(SHs)dSHs+H(222H1)t0f(SHs)s2H1ds

for all t ∈ [0, T].

3 Some Basic Estimates

Throughout this paper we assume that θ < 0 and 12<H<1. Recall that the linear self-interacting diffusion with sub-fBm SH defined by the stochastic differential equation

XHt=SHtθt0s0(XHsXHu)duds+νt,t0

with θ < 0. Define the kernel (t, s)↦hθ(t, s) as follows

hθ(t,s)={1θse12θs2tse12θu2du,ts,0,t<s

for s, t ≥ 0. By the variation of constants method (see, Cranston and Le Jan [1]) or Itô’s formula we may introduce the following representation:

XHt=t0hθ(t,s)dSHs+νt0hθ(t,s)ds

for t ≥ 0.

The kernel function (t, s)↦hθ(t, s) with θ < 0 admits the following properties (these properties are proved partly in Sun and Yan [12]):

• For all s ≥ 0, the limit

limt(te12θt2hθ(t,s))=se12θs2

for all s ≥ 0.

• For all ts ≥ 0, we have

1hθ(t,s)e12θ(t2s2).

• For all ts, r ≥ 0, we have

hθ(t,0)=hθ(t,t)=1,tshθ(t,u)du=e12θs2tse12θu2du.

Lemma 3.1 Let θ < 0 and define function

Iθ(t)=θte12θt2t0e12θu2du1.

We then have limtt2Iθ(t)=1θ and

limtt2(1+θte12θt2te12θu2du)=1θ

Proof This is simple calculus exercise.

Lemma 3.2 (Sun and Yan [12]). Let θ < 0 and define the functions tIθ(t, n), n = 1, 2, … as follows

Iθ(t,1)=θt2Iθ(t),Iθ(t,n+1)=θt2[Iθ(t,n)(2n1)].

Then we have

limtIθ(t,n)=(2n1).

for every n ≥ 0, where (−1)! = 1.

Lemma 3.3 Let θ < 0. Then the integral

Δ(H)=00xye12θ(x2+y2)ψH(x,y)dxdy

converges and as t,

limtt2eθt2E(XHt)2=Δ(H).

Proof An elementary may show that (3.6) converges for all θ < 0. It follows from L’Hôspital’s rule that

limtt2eθt2E(XHt)2=limtt2eθt2t0t0hθ(t,x)hθ(t,y)ψH(x,y)dxdy=limtθ2t2eθt2t0dxt0xye12θ(x2+y2)ψH(x,y)dytxdutye12θ(u2+v2)dv=2limtθ2t2eθt2t0duu0dxu0dvv0dyxye12θ(x2+y2u2v2)ψH(x,y)=limtθt1e12θt2t0dxt0e12θv2dvv0xye12θ(x2+y2)ψH(x,y)dy=limtθt1e12θt2t0e12θv2dvt0dxv0xye12θ(x2+y2)ψH(x,y)dy=limtθt1e12θt2t0e12θv2dvv0dxv0xye12θ(x2+y2)ψH(x,y)dy=0dx0xye12θ(x2+y2)ψH(x,y)dy,

where we have used the following fact:

limt1t1e12θt2t0e12θv2dvtvdxv0xye12θ(x2+y2)ψH(x,y)dy=limt1t1e12θt2t0dxx0e12θv2dvv0xye12θ(x2+y2)ψH(x,y)dy=0.

This completes the proof.

Lemma 3.4 Let θ < 0. Then, convergence

limt1t22Heθt2tssre12θ(s2+r2)ψH(s,r)dsdr=14(θ)2HΓ(2H+1).

holds.

Proof It follows from L’Hôspital’s rule that

limt1t22Heθt2tue12θu2(uve12θv2ψH(u,v)dv)du=12θlimt1t22He12θt2tve12θv2ψH(t,v)dv=limtH(2H1)2θt22Htve12θ(v2t2)((vt)2H2(v+t)2H2)dv

for all θ < 0 and 12<H<1. By making the change of variable 12θ(v2t2)=x, we see that

limt12θt22Htve12θ(v2t2)((vt)2H2(v+t)2H2)dv=limt12θ2t22H0ex{(t2+2xθt)2H2(t2+x+t)2H2}dx=limt12θ2t22H0ex(2xθ)2H2(t2+2xθ+t)22Hdxlimt12θ2t22H0ex(t2+x+t)2H2dx=12(θ)2H1Γ(2H1)

for all θ < 0 and 12<H<1. This completes the proof.

Lemma 3.5 Let θ < 0 and 0 ≤ s < tT. We then have

c(ts)2HE[(XHtXHs)2]C(ts)2H

Proof Given 0 ≤ s < tT and denote

ˆXHt=t0hθ(t,r)dSHr,t0.

It follows that

E[(ˆXHtˆXHs)2]=E(s0[hθ(t,x)hθ(s,x)]dSHx)2+E(tshθ(t,x)dSHx)2+2E(tshθ(t,y)dSHys0[hθ(t,x)hθ(s,x)]dSHx).

Now, we estimate the three terms. For the first term, we have

0E(s0[hθ(t,x)hθ(s,x)]dSHx)2=s0s0(hθ(t,x)hθ(s,x))(hθ(t,y)hθ(s,y))ψH(x,y)dxdy=θ2(tse12θu2du)2s0s0xye12θ(x2+y2)ψH(x,y)dxdyθ2s2(ts)2eθt2s0s0ψH(x,y)dxdy=θ2s2(ts)2eθt2E(SHs)2CH,T(ts)2

for all θ < 0 and 0 < s < tT. For the second term, we have

E(tshθ(t,x)dSHx)2=tstshθ(t,x)hθ(t,y)ψH(x,y)dxdyeθt2tstsxye12θ(x2+y2)ψH(x,y)dxdyt2eθt2tstsψH(x,y)dxdyCH,T(ts)2H.

for all θ < 0 and 0 < s < tT. Similarly, for the third term, we also prove

0E(tshθ(t,y)dSHys0[hθ(t,x)hθ(s,x)]dSHx)=tss0hθ(t,y)[hθ(t,x)hθ(s,x)]ψH(x,y)dxdyθ2e12θt2(tse12θu2du)tsye12θy2dys0xe12θx2ψH(x,y)dxθ2eθt2(ts)tsye12θy2dys0xe12θx2ψH(x,y)dxCH,T(ts)2

for all θ < 0 and 0 < s < tT. Thus, we have obtained the following estimate:

E[(ˆXHtˆXHs)2]CH,T|ts|2H

for all θ < 0 and 0 < s < tT.On the other hand, elementary calculations may show that

s0[hθ(t,r)hθ(s,r)]dr=θtse12θu2dus0re12θr2drCH,T(ts)

and

tshθ(t,r)dr=e12θs2tse12θr2drCH,T(ts)

for all θ < 0 and 0 < s < tT. It follows that

(t0hθ(t,r)drs0hθ(s,r)dr)2=(s0[hθ(t,r)hθ(s,r)]dr)2+(tshθ(t,r)dr)2+2tshθ(t,r)drs0[hθ(t,r)hθ(s,r)]drCH,T(ts)2

for all θ < 0 and 0 < s < tT, which implies that

E[(Xa,btXa,bs)2]=E[(ˆXa,btˆXa,bs)2]+ν2(t0hθ(t,r)drs0hθ(s,r)dr)2CH,T(ts)2H

for all θ < 0 and 0 < s < tT. Noting that the above calculations are invertible for all θ < 0 and 0 < s < tT, one can obtain the left hand side in (3.8) and the lemma follows.

4 Convergence

In this section, we obtain the large time behaviors associated with the solution XH to Eq. 3.1. From Lemma 3.5 and Guassianness, we find that the self-repelling diffusion {XHt,t0} is H-Hölder continuous. So, the integral

t0sdXHs

exists with t ≥ 0 as a Young integral and

tXHt=t0sdXHs+t0XHsds

for all t ≥ 0. Define the process Y = {Yt, t ≥ 0} by

Yt:=t0(XHtXHs)ds=tXHtt0XHsds=t0sdXHs=t0sdSHst0θsYsds+12νt2.

By the variation of constants method, one can prove

Yt=e12θt2t0se12θs2dSHsνθ(e12θt21)

for all t ≥ 0. Define Gaussian process ξH={ξHt,t0} as follows

ξHtt0se12θs2dSHs,t0.

Lemma 4.1 Let θ < 0 and 12<H<1. Then, the random variable

ξH0se12θs2dSHs

exists as an element in L2. Moreover, ξH is H-Hölder continuous and ξHtξH in L2 and almost surely, as t tends to infinity.

Proof This is simple calculus exercise. In fact, we have

E(0xe12θx2dSHx)2=00xye12θ(x2+y2)ψH(x,y)dxdy=20xe12θx2dxx0ye12θy2ψH(x,y)dy=2H(2H1)0xe12θx2dxx0ye12θy2((xy)2H2(x+y)2H2)dy2H(2H1)0xe12θx2dxx0((xy)2H2(x+y)2H2)ydy=2H(2H1)CH0x2H+1e12θx2dx=Cθ,HΓ(2H+2)

for all θ < 0 and 12<H<1, which shows that the random variable ξH exists as an element in L2.

Now, we show that the process ξa,b is Hölder continuous. For all 0 < s < t by the inequality ex2xC for all x ≥ 0, we have

E(ξHtξHs)2=E(tsxe12θx2dSHx)2=tstsxye12θ(x2+y2)ψH(x,y)dxdy=2tsxe12θx2dxxsye12θy2ψH(x,y)dy=2H(2H1)tsxe12θx2dxxsye12θy2((xy)2H2(x+y)2H2)dy2HCθ(2H1)tsdxxs(xy)2H2dy=Cθ,H(ts)2H.

Thus, the normality of ξH implies that

E(ξHtξHs)2nCθ,H,n(ts)2nH

for all 0 ≤ s < t, 12<H<1 and integer numbers n ≥ 1, and the Hölder continuity follows.

Nextly, we check the ξa,bt converges to ξH in L2. This follows from the next estimate:

E(ξHtξH)2=ttxye12θ(x2+y2)ψH(x,y)dxdy=2txtxye12θ(x2+y2)ψH(x,y)dxdy2e12θt2txe12θx2dxxtyψH(x,y)dy2e12θt2txe12θx2dxx0yψH(x,y)dy2H(2H1)e12θt2txe12θx2dxx0y((xy)2H2(x+y)2H2)dy2H(2H1)e12θt2txe12θx2dxx0y(xy)2H2dy=2H(2H1)(10u(1u)2H2du)e12θt2tx2H+1e12θx2dx0,

as t tends to infinity.

Finally, we check the ξa,bt converges to ξH almost surely. By integration by parts we see that

ξHtξH=tse12θs2dSHs=te12θt2SHtt(1+θs2)e12θs2SHsds

for all t ≥ 0. Elementary may check that the convergence

ηHtt(1+θs2)e12θs2SHsdsa.s0

holds almost surely, as t tends to infinity. In fact, by inequality

tsαe12θs2dsCtα1e12θt2,α>1,

with t ≥ 0, we may show that

E(supnt<n+1|ηHt|2)nn(1+θs2)(1+θr2)e12θ(s2+r2)E|SHsSHr|drdsC(ns2+He12θs2ds)2Cn2+2Heθn2,

for all integer numbers n ≥ 1, and hence

n=0P(supnt<n+1|ηHt|2ε)Cε2n=0n2+2Heθn2<.

Thus, Borel-Cantelli’s lemma implies that ηHt converges to zero almost surely as t tends to infinity, and the lemma follows from (4.2).

Corollary 4.1 For all γ > 0, we have

tγ(ξHtξH)=tγtse12θs2dSHs0,

in L2 and almost surely, as t tends to infinity.

Lemma 4.2 Let θ < 0 and 12<H<1. Then, we have

Λγ(t,θ)tγ+1e12θt2t0e12θu2(ξa,bξa,bu)du0

in L2 and almost surely for every γ ≥ 0, as t tends to infinity.

Proof Given 0 < st, θ < 0 and denote

ϒθ(s,t)t0e12θv2dvvre12θr2ψH(s,r)dr=t0re12θr2ψH(s,r)drr0e12θv2dv+(t0e12θv2dv)tre12θr2ψH(s,r)drCt0rψH(s,r)dr+Cte12θt2tre12θr2ψH(s,r)drC(t0rψH(s,r)dr+(ts)2H2t1),

where we have used the fact

x0e12θv2dvCxe12θx2,x0

and estimates

tre12θr2ψH(s,r)dr=H(2H1)tr((rs)2H2(s+r)2H2)e12θr2drH(2H1)tr(rs)2H2e12θr2drH(2H1)(ts)2H2tre12θr2dr=H(2H1)θ(ts)2H2e12θt2.

It follows that

E|Λγ(t,θ)|2=t2γ+2eθt2t0t0e12θ(u2+v2)E(use12θs2dSHs)(vre12θr2dSHr)dudv=t2γ+2eθt2t0t0e12θ(u2+v2)dudvuvrse12θ(r2+s2)ψH(s,r)drds=t2γ+2eθt2t0e12θu2duuse12θs2ϒθ(s,t)ds=t2γ+2eθt2t0se12θs2ψH(s,θ)dss0e12θu2du+t2γ+2eθt2tse12θs2ϒθ(s,t)dst0e12θu2dut2γ+2eθt2t0s2ϒθ(s,t)ds+t2γ+1e12θt2tse12θs2ϒθ(s,t)ds0(t),

which shows that Λγ(t, θ) converges to zero in L2.Now, we obtain the convergence with probability one. Noting that

ξHξHu=use12θs2dSHs

for all u ≥ 0, we get

|Λγ(t,θ)|tγ+1e12θt2t0e12θu2|use12θs2dSHs|dutγ+1e12θt2t0e12θu2(u|SHu|e12θu2+u|SHs(1θs2)|e12θs2ds)du=tγ+1e12θt2t0u|SHu|du+tγ+1e12θt2t0e12θu2duu|SHs(1θs2)|e12θs2ds=tγ+1e12θt2t0u|SHu|du+tγ+1e12θt2t0|SHs(1θs2)|e12θs2dss0e12θu2du+tγ+1e12θt2t|SHs(1θs2)|e12θs2dst0e12θu2dutγ+1e12θt2t0u|SHu|du+tγ+1e12θt2t0|SHs(1θs2)|sds+Cθtγt|SHs(1θs2)|e12θs2ds0

almost surely for all γ ≥ 0, θ < 0 and 12<H<1, as t tends to infinity. This completes the proof.

The objects of this paper are to prove the following theorems which give the long time behaviors for XH with 12<H<1.

Theorem 4.1 Let θ < 0 and 12<H<1. Then, as t∞, the convergence

J0Ht;θ,νte12θt2XtHξHνθ

holds in L2 and almost surely.

Proof Given t > 0 and θ < 0. Simple calculations may prove

J0Ht;θ,ν=te12θt2XtH=te12θt20thθt,sdSsH+νte12θt20thθt,sds=te12θt2StHθt2e12θt20tse12θs2ste12θu2dudSsH+νte12θt20te12θs2ds=te12θt2StHθte12θt20te12θu20use12θs2dSsHdu+νte12θt20te12θs2ds=te12θt2StHθte12θt20te12θu2ξuHdu+νte12θt20te12θs2ds.

It follows from Lemma 4.1, Corollary 4.1, and Lemma 4.2 that

J0Ht;θ,νξHνθ=te12θt2XtHξHνθ=te12θt2StHθte12θt20te12θu2ξuHξHdu+ξHνθθte12θt20te12θu2du10t

in L2 and almost surely for all θ < 0 and 12<H<1, as t tends to infinity.

Theorem 4.2 Define the processes JH(n,θ,ν)={JtH(n,θ,ν),t0},n1 by

JnHt;θ,νθt2Jn1Ht;θ,ν2n3ξHνθ,n=1,2,,

for all t ≥ 0, where (−1)!! = 1. Then, the convergence

JnHt;θ,ν2n1ξHνθ

holds in L2 and almost surely for every n ≥ 1, as t∞.

Proof From the proof of Theorem 4.1, we find that the identities

J0Ht;θ,νξHνθ=te12θt2StH+θte12θt20te12θu2ξuHξHdu+ξHνθθte12θt20te12θu2du1,JnHt;θ,ν=ξHνθInt,θ+tθt2ne12θt2StH+θtθt2ne12θt20te12θu2ξuHξHdu.

holds for all t > 0, n ≥ 1 and θ < 0, where In(t, θ) is given in Lemma 3.2. Thus, the theorem follows from Lemma 4.1, Corollary 4.1, Lemma 4.2 and Theorem 4.1.

5 Simulation

We have applied our results to the following linear self-repelling diffusion driven by a sub-fBm SH with 12<H<1:

dXtH=dStHθ0tXtHXsHdsdt+νdt,X0H=0,

where θ < 0 and νR are two parameters. We will simulate the process with ν = 0 in the following cases:

H = 0.7 and θ = − 1, θ = − 10, and θ = − 100, respectively (see, Figure 1, Figure 2, Figure 3, and Table 1, Table 2, Table 3);

H = 0.5 and θ = − 1, θ = − 10, and θ = − 100, respectively (see, Figure 4, Figure 5, Figure 6, and Table 4, Table 5, Table 6);

Remark 1 From the following numerical results, we can find that it is important to study the estimates of parameters θ and ν.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Author Contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

Funding

This study was funded by the National Natural Science Foundation of China (NSFC), grant no. 11971101.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

1. Cranston M, Le Jan Y. Self Attracting Diffusions: Two Case Studies. Math Ann (1995) 303:87–93. doi:10.1007/bf01460980

CrossRef Full Text | Google Scholar

2. Durrett RT, Rogers LCG. Asymptotic Behavior of Brownian Polymers. Probab Th Rel Fields (1992) 92:337–49. doi:10.1007/bf01300560

CrossRef Full Text | Google Scholar

3. Pemantle R. Phase Transition in Reinforced Random Walk and RWRE on Trees. Ann Probab (1988) 16:1229–41. doi:10.1214/aop/1176991687

CrossRef Full Text | Google Scholar

4. Benaïm M, Ledoux M, Raimond O. Self-interacting Diffusions. Probab Theor Relat Fields (2002) 122:1–41. doi:10.1007/s004400100161

CrossRef Full Text | Google Scholar

5. Benaïm M, Ciotir I, Gauthier C-E. Self-repelling Diffusions via an Infinite Dimensional Approach. Stoch Pde: Anal Comp (2015) 3:506–30. doi:10.1007/s40072-015-0059-5

CrossRef Full Text | Google Scholar

6. Cranston M, Mountford TS. The strong Law of Large Numbers for a Brownian Polymer. Ann Probab (1996) 24:1300–23. doi:10.1214/aop/1065725183

CrossRef Full Text | Google Scholar

7. Gauthier C-E. Self Attracting Diffusions on a Sphere and Application to a Periodic Case. Electron Commun Probab (2016) 21(No. 53):1–12. doi:10.1214/16-ecp4547

CrossRef Full Text | Google Scholar

8. Herrmann S, Roynette B. Boundedness and Convergence of Some Self-Attracting Diffusions. Mathematische Annalen (2003) 325:81–96. doi:10.1007/s00208-002-0370-0

CrossRef Full Text | Google Scholar

9. Herrmann S, Scheutzow M. Rate of Convergence of Some Self-Attracting Diffusions. Stochastic Process their Appl (2004) 111:41–55. doi:10.1016/j.spa.2003.10.012

CrossRef Full Text | Google Scholar

10. Mountford T, Tarrés P. An Asymptotic Result for Brownian Polymers. Ann Inst H Poincaré Probab Statist (2008) 44:29–46. doi:10.1214/07-aihp113

CrossRef Full Text | Google Scholar

11. Shen L, Xia X, Yan L. Least Squares Estimation for the Linear Self-Repelling Diffusion Driven by α-stable Motions. Appear Statist Probab Lett (2021) 181:109259. doi:10.1016/j.spl.2021.109259

CrossRef Full Text | Google Scholar

12. Sun X, Yan L. A Convergence on the Linear Self-Interacting Diffusion Driven by α-stable Motion. Stochastics 93(2021):1186–1208. doi:10.1080/17442508.2020.1869239

CrossRef Full Text | Google Scholar

13. Yan L, Sun Y, Lu Y. On the Linear Fractional Self-Attracting Diffusion. J Theor Probab (2008) 21:502–16. doi:10.1007/s10959-007-0113-y

CrossRef Full Text | Google Scholar

14. Sun X, Yan L. The Laws of Large Numbers Associated with the Linear Self-Attracting Diffusion Driven by Fractional Brownian Motion and Applications. J Theoret Prob (2021). [Epub of Print]. doi:10.1007/s10959-021-01126-0

CrossRef Full Text | Google Scholar

15. Gan Y, Yan L. Least Squares Estimation for the Linear Self-Repelling Diffusion Driven by Fractional Brownian Motion (In Chinese). Sci CHINA Math (2018) 48:1143–58. doi:10.1360/scm-2017-0387

CrossRef Full Text | Google Scholar

16. Alós E, Mazet O, Nualart D. Stochastic Calculus with Respect to Gaussian Processes. Ann Prob (2001) 29:766–801. doi:10.1214/aop/1008956692

CrossRef Full Text | Google Scholar

17. Bojdecki T, Gorostiza LG, Talarczyk A. Some Extensions of Fractional Brownian Motion and Sub-fractional Brownian Motion Related to Particle Systems. Elect Comm Probab (2007) 12:161–72. doi:10.1214/ecp.v12-1272

CrossRef Full Text | Google Scholar

18. Bojdecki T, Gorostiza LG, Talarczyk A. Sub-fractional Brownian Motion and its Relation to Occupation Times. Stat Probab Lett (2004) 69:405–19. doi:10.1016/j.spl.2004.06.035

CrossRef Full Text | Google Scholar

19. Bojdecki T, Gorostiza LG, Talarczyk A. Occupation Time Limits of Inhomogeneous Poisson Systems of Independent Particles. Stochastic Process their Appl (2008) 118:28–52. doi:10.1016/j.spa.2007.03.008

CrossRef Full Text | Google Scholar

20. Bojdecki T, Gorostiza LG, Talarczyk A. Self-Similar Stable Processes Arising from High-Density Limits of Occupation Times of Particle Systems. Potential Anal (2008) 28:71–103. doi:10.1007/s11118-007-9067-z

CrossRef Full Text | Google Scholar

21. Li M. Modified Multifractional Gaussian Noise and its Application. Phys Scr (2021) 96(1212):125002. doi:10.1088/1402-4896/ac1cf6

CrossRef Full Text | Google Scholar

22. Li M. Generalized Fractional Gaussian Noise and its Application to Traffic Modeling. Physica A 579(202122):1236137. doi:10.1016/j.physa.2021.126138

CrossRef Full Text | Google Scholar

23. Li M. Multi-fractional Generalized Cauchy Process and its Application to Teletraffic. Physica A: Stat Mech its Appl (2020) 550(14):123982. doi:10.1016/j.physa.2019.123982

CrossRef Full Text | Google Scholar

24. Li M. Fractal Time Series-A Tutorial Review. Math Probl Eng (2010) 2010:26. Article ID 157264. doi:10.1155/2010/157264

CrossRef Full Text | Google Scholar

25. Shen G, Yan L. An Approximation of Subfractional Brownian Motion. Commun Stat - Theor Methods (2014) 43:1873–86. doi:10.1080/03610926.2013.769598

CrossRef Full Text | Google Scholar

26. Shen G, Yan L. Estimators for the Drift of Subfractional Brownian Motion. Commun Stat - Theor Methods (2014) 43:1601–12. doi:10.1080/03610926.2012.697243

CrossRef Full Text | Google Scholar

27. Sun X, Yan L. A central Limit Theorem Associated with Sub-fractional Brownian Motion and an Application (In Chinese). Sci Sin Math (2017) 47:1055–1076. doi:10.1360/scm-2016-0748

CrossRef Full Text | Google Scholar

28. Tudor C. Some Properties of the Sub-fractional Brownian Motion. Stochastics (2007) 79:431–48. doi:10.1080/17442500601100331

CrossRef Full Text | Google Scholar

29. Tudor C. Inner Product Spaces of Integrands Associated to Subfractional Brownian Motion. Stat Probab Lett (2008) 78:2201–9. doi:10.1016/j.spl.2008.01.087

CrossRef Full Text | Google Scholar

30. Tudor C. On the Wiener Integral with Respect to a Sub-fractional Brownian Motion on an Interval. J Math Anal Appl (2009) 351:456–68. doi:10.1016/j.jmaa.2008.10.041

CrossRef Full Text | Google Scholar

31. Tudor C. Some Aspects of Stochastic Calculus for the Sub-fractional Brownian Motion. Ann Univ Bucuresti Mathematica (2008) 24:199–230.

Google Scholar

32. Ciprian A. Tudor, Analysis Of Variations For Self-Similar Processes. Heidelberg, New York: Springer (2013).

Google Scholar

33. Yan L, He K, Chen C. The Generalized Bouleau-Yor Identity for a Sub-fractional Brownian Motion. Sci China Math (2013) 56:2089–116. doi:10.1007/s11425-013-4604-2

CrossRef Full Text | Google Scholar

34. Yan L, Shen G. On the Collision Local Time of Sub-fractional Brownian Motions. Stat Probab Lett (2010) 80:296–308. doi:10.1016/j.spl.2009.11.003

CrossRef Full Text | Google Scholar

35. Yan L, Shen G, He K. Itô’s Formula for the Sub-fractional Brownian Motion. Comm Stochastic Anal (2011) 5:135–59. doi:10.31390/cosa.5.1.09

CrossRef Full Text | Google Scholar

36. Nualart D. Malliavin Calculus and Related Topics. 2nd ed. New York: Springer (2006).

Google Scholar

37. Bertoin J. Sur une intégrale pour les processus á α-variation borné. Ann Probab (1989) 17:1521–35. doi:10.1214/aop/1176991171

CrossRef Full Text | Google Scholar

38. FöIllmer H, Calcul d’Itô Sans Probabilités, Azéme and Yor : Séinaire de Probabilités XV, Lecture Notes in Math, No 50. Springer-Verlag (1980). p. 143–5.

Google Scholar

Keywords: the self-repelling diffusion, asymptotic distribution, convergence, sub-fractional Brownian motion, stochastic integral

Citation: Gao H, Guo R, Jin Y and Yan L (2022) Large Time Behavior on the Linear Self-Interacting Diffusion Driven by Sub-Fractional Brownian Motion With Hurst Index Large Than 0.5 I: Self-Repelling Case. Front. Phys. 9:795210. doi: 10.3389/fphy.2021.795210

Received: 14 October 2021; Accepted: 05 November 2021;
Published: 14 January 2022.

Edited by:

Ming Li, Zhejiang University, China

Reviewed by:

Yu Sun, Our Lady of the Lake University, United States
Zhenxia Liu, Linköping University, Sweden

Copyright © 2022 Gao, Guo, Jin and Yan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Han Gao, 1061760802@qq.com

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.