Analysis of the fractional diarrhea model with Mittag-Leffler kernel

Muhammad Sajid Iqbal; Nauman Ahmed; Ali Akgül; Ali Raza; Muhammad Shahzad; Zafar Iqbal; Muhammad Rafiq; Fahd Jarad; Muhammad Sajid Iqbal; Nauman Ahmed; Ali Akgül; Ali Raza; Muhammad Shahzad; Zafar Iqbal; Muhammad Rafiq; Fahd Jarad

doi:10.3934/math.2022720

AIMS Mathematics

2022, Volume 7, Issue 7: 13000-13018. doi: 10.3934/math.2022720

Previous Article Next Article

Research article Special Issues

Analysis of the fractional diarrhea model with Mittag-Leffler kernel

1.
Department of Humanities and Basic Sciences, MCS, National University of Sciences and Technology, Islamabad, Pakistan
2.
Department of Mathematics and Statistics, The University of Lahore, Lahore, Pakistan
3.
Siirt University, Art and Science Faculty, Department of Mathematics, 56100 Siirt, Turkey
4.
Department of Mathematics, Govt. Maulana Zafar Ali Khan Graduate College Wazirabad, 52000, Punjab Higher Education Department (PHED), Lahore, 54000, Pakistan
5.
Department of Mathematics, University of Management and Technology, Lahore, Pakistan
6.
Department of Mathematics, Faculty of Science, University of Central Punjab, Lahore, Pakistan
7.
Department of Mathematics, Çankaya University, Etimesgut 06790, Ankara, Turkey
8.
Department of Mathematics, King Abdulaziz University, Jeddah, Saudi Arabia
9.
Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan

Received: 21 December 2021 Revised: 03 April 2022 Accepted: 14 April 2022 Published: 09 May 2022
MSC : 65P99, 92D25

In this article, we have introduced the diarrhea disease dynamics in a varying population. For this purpose, a classical model of the viral disease is converted into the fractional-order model by using Atangana-Baleanu fractional-order derivatives in the Caputo sense. The existence and uniqueness of the solutions are investigated by using the contraction mapping principle. Two types of equilibrium points i.e., disease-free and endemic equilibrium are also worked out. The important parameters and the basic reproduction number are also described. Some standard results are established to prove that the disease-free equilibrium state is locally and globally asymptotically stable for the underlying continuous system. It is also shown that the system is locally asymptotically stable at the endemic equilibrium point. The current model is solved by the Mittag-Leffler kernel. The study is closed with constraints on the basic reproduction number $R_{0}$ and some concluding remarks.

Keywords:

Citation: Muhammad Sajid Iqbal, Nauman Ahmed, Ali Akgül, Ali Raza, Muhammad Shahzad, Zafar Iqbal, Muhammad Rafiq, Fahd Jarad. Analysis of the fractional diarrhea model with Mittag-Leffler kernel[J]. AIMS Mathematics, 2022, 7(7): 13000-13018. doi: 10.3934/math.2022720

Related Papers:

[1]	Muhammad Qiyas, Talha Madrar, Saifullah Khan, Saleem Abdullah, Thongchai Botmart, Anuwat Jirawattanapaint . Decision support system based on fuzzy credibility Dombi aggregation operators and modified TOPSIS method. AIMS Mathematics, 2022, 7(10): 19057-19082. doi: 10.3934/math.20221047
[2]	Muhammad Qiyas, Neelam Khan, Muhammad Naeem, Saleem Abdullah . Intuitionistic fuzzy credibility Dombi aggregation operators and their application of railway train selection in Pakistan. AIMS Mathematics, 2023, 8(3): 6520-6542. doi: 10.3934/math.2023329
[3]	Shichao Li, Zeeshan Ali, Peide Liu . Prioritized Hamy mean operators based on Dombi t-norm and t-conorm for the complex interval-valued Atanassov-Intuitionistic fuzzy sets and their applications in strategic decision-making problems. AIMS Mathematics, 2025, 10(3): 6589-6635. doi: 10.3934/math.2025302
[4]	Attaullah, Shahzaib Ashraf, Noor Rehman, Asghar Khan, Muhammad Naeem, Choonkil Park . Improved VIKOR methodology based on $q$ -rung orthopair hesitant fuzzy rough aggregation information: application in multi expert decision making. AIMS Mathematics, 2022, 7(5): 9524-9548. doi: 10.3934/math.2022530
[5]	Muhammad Qiyas, Muhammad Naeem, Neelam Khan, Lazim Abdullah . Bipolar complex fuzzy credibility aggregation operators and their application in decision making problem. AIMS Mathematics, 2023, 8(8): 19240-19263. doi: 10.3934/math.2023981
[6]	Saleem Abdullah, Muhammad Qiyas, Muhammad Naeem, Mamona, Yi Liu . Pythagorean Cubic fuzzy Hamacher aggregation operators and their application in green supply selection problem. AIMS Mathematics, 2022, 7(3): 4735-4766. doi: 10.3934/math.2022263
[7]	Attaullah, Shahzaib Ashraf, Noor Rehman, Asghar Khan, Choonkil Park . A decision making algorithm for wind power plant based on q-rung orthopair hesitant fuzzy rough aggregation information and TOPSIS. AIMS Mathematics, 2022, 7(4): 5241-5274. doi: 10.3934/math.2022292
[8]	Rana Muhammad Zulqarnain, Xiao Long Xin, Muhammad Saeed . Extension of TOPSIS method under intuitionistic fuzzy hypersoft environment based on correlation coefficient and aggregation operators to solve decision making problem. AIMS Mathematics, 2021, 6(3): 2732-2755. doi: 10.3934/math.2021167
[9]	Esmail Hassan Abdullatif Al-Sabri, Muhammad Rahim, Fazli Amin, Rashad Ismail, Salma Khan, Agaeb Mahal Alanzi, Hamiden Abd El-Wahed Khalifa . Multi-criteria decision-making based on Pythagorean cubic fuzzy Einstein aggregation operators for investment management. AIMS Mathematics, 2023, 8(7): 16961-16988. doi: 10.3934/math.2023866
[10]	Qiansheng Zhang, Jingfa Liu . Entropy of credibility distribution for intuitionistic fuzzy variable. AIMS Mathematics, 2023, 8(4): 9671-9691. doi: 10.3934/math.2023488

Abstract

1. Introduction

Stochastic differential equations (SDEs) are used to describe phenomena in the context of natural science and engineering. To consider models of phenomena whose future states depend not only on the present states but also on the past states, stochastic differential delay equations (SDDEs) are used. For example, we refer to ^[1], in which the delay market model is studied. The study of SDDEs was started in a previous paper ^[2], in which Ito and Nisio considered SDEs that depend on infinite past processes. Following the paper, many properties of SDDEs have been discovered. Refer to the paper by Ivanov et al. ^[3] for a survey.

Because solutions of SDDEs are influenced by past events, they do not have Markov properties. This makes representations of solutions complicated; therefore, approximate solutions of SDDEs have been studied.

In ^[4], the strong discrete time approximation of an SDDE with a single constant time delay is studied. Under the global Lipschitz condition, the explicit solutions for linear stochastic delay equations are given. The global Lipschitz condition has been relaxed to the local Lipschitz condition in ^[5] and ^[6]. However, studies of SDDEs with a single constant time delay are still ongoing under the global Lipschitz condition (cf. ^[7] and ^[8]).

As mentioned in ^[9], when we consider approximate solutions of SDEs, it is important to know the error rate of convergence of approximate solutions to the exact solutions. In ^[10] and ^[11], Kanagawa obtained the rate of convergence of Euler–Maruyama approximate solutions of SDEs in the $L^2$ -mean and $L^p$ -mean for some $p \geq 2$ , respectively. In ^[12], we stated the rate of convergence in the $L^2$ -mean for SDDEs with a single delay function. In ^[13], we studied the cases of SDDEs with multiple delay functions and stated the rate of convergence in the $L^2$ -mean. However, in ^[12] and ^[13], the results were given without detail of proof.

The Euler–Maruyama approximation scheme for SDEs is implemented by a step function, which is a discretization of Brownian motion $B(t)$ . The random increments are given by $\varDelta B_n = B(t_n)-B(t_{n-1}), n = 1, 2, \ldots$ , and the values provided by pseudo-random variables. Brownian motion, however, moves the time interval $[t_{n-1}, t_n]$ . With the awareness of such issues, Kanagawa proposed a method of confidence interval estimations to predict the exact solutions in ^[14]. His method is supported by stochastic analysis (^[10] and ^[11]) and Chebyshev's inequality. Through confidence interval estimations, we can expect sample paths of solutions of SDEs in each time interval (see Figure 6 in ^[9] and ^[15]). We remark that it is not possible to obtain confidence intervals by generating many trajectories using simulation studies and measuring the results, because we cannot obtain information on the behavior of $\{B(t), t \in (t_{n-1}, t_n)\}$ . Refer to Chapter 11 in ^[16], ^[9], ^[17] and ^[18] for details on simulation studies for SDEs. In the case of SDDEs, the confidence interval estimations for the Euler–Maruyama approximate solutions were studied in ^[12] and ^[13] only for the special case of SDDEs, which have no drift terms following ^[14] and ^[15].

This paper is a continuation of our previous works ^[12] and ^[13]. In this paper, we consider SDDEs under the global Lipschitz condition containing multiple delay functions. In this sense, the model considered in this paper is an extension of those in ^[4], ^[7] and ^[8]. The purpose of this paper is two-fold. The first is to state a convergence theorem in the $L^p$ -mean for some $p \geq 2$ following the method by Kanagawa (^[11]). We remark that the convergence theorem showed in this paper includes the results in ^[12] and ^[13]. The second is to present the confidence interval estimations for the general case of SDDEs, which have diffusion terms and drift terms.

This paper is organized as follows. In the following section, we provide a setting of the SDDE and its Euler–Maruyama approximation scheme. After that we state our main theorem, i.e., the rate of convergence in the $L^p$ -mean for some $p \geq 2$ . In Section 3, we provide confidence interval estimations for the Euler–Maruyama approximate solutions of SDDEs for the general case and the special case. In Section 4, we present numerical examples of confidence interval estimations provided in Section 3. In Section 5, we provide proofs for the general case of SDDEs. We also provide the proofs for the special case of SDDEs.

2. Setting of SDDE and its solution

2.1. Setting of SDDE

Let $(\Omega, {\cal F}, P)$ be a complete probability space. On the space, we provide a filtration $\{{\cal F}_t\}_{t \geq 0}$ that satisfies the usual conditions; that is, it is right continuous and ${\cal F}_0$ contains all $P$ -null sets. We denote such a space by $(\Omega, {\cal F}, P; {\cal F}_t)$ . Let $B(t) = {}^t(B_1(t), \ldots, B_m(t))$ be an $m$ -dimensional standard Brownian motion on $(\Omega, {\cal F}, P; {\cal F}_t)$ . Let $\tau > 0$ and $C([-\tau, 0]; {\mathbb R^n})$ denotes the space of continuous functions $\xi: [-\tau, 0] \to {\mathbb R}^n$ with norms generated by $\sup_{-\tau \leq t \leq 0}|\xi(t)|$ . Additionally, $C_{{\mathcal F}_t}([a, b]; {\mathbb R}^n)$ denotes the family of ${\mathcal F}_t$ -measurable $C([a, b]; {\mathbb R}^n)$ -valued random variables.

We next provide a setting of an SDDE. We first introduce delay functions as follows:

(D) (ⅰ) Let $\delta_i(t)$ be a Borel measurable function such that

$\begin{align} -\tau \leq \delta_i(t) \leq t\ \mbox{for}\ i = 1, \ldots, \ell. \end{align}$

(2.1)

(ⅱ) For each $\delta_i$ , we assume that

$\begin{align} |\delta_{i}(t)-\delta_{i}(s)|\leq \rho |t-s|, \ s, t\geq 0 \ \mbox{for some positive constant}\ \rho. \end{align}$

(2.2)

Initial data are given by information for $t \leq 0$ , which is denoted by $\{\xi(t), -\tau \leq t \leq 0\} \in C_{{\mathcal F}_0}([-\tau, 0]; {\mathbb R}^n)$ . For $\xi$ , we assume the following:

(P) (ⅰ) For some $p \geq 2$ , there exists $K_0 < \infty$ such that

$\begin{align} \sup\limits_{-\tau \leq t \leq 0} E[|\xi(t)|^p] = K_0. \end{align}$

(2.3)

(ⅱ) For the same $p$ in (2.3), there exist $K_1 > 0$ and $\gamma \in (0, 1]$ such that

$\begin{align} E[|\xi(t)-\xi(s)|^p]\leq K_{1}(t-s)^\gamma, \ -\tau \leq s < t \leq 0. \end{align}$

(2.4)

Let $f(x_0, \ldots, x_{\ell})$ and $g(x_0, \ldots, x_{\ell})$ be ${\mathcal B}({\mathbb R^n}) \otimes {\mathcal B}({\mathbb R^{n \times \ell}})$ -measurable functions with values in ${\mathbb R}^n$ and ${\mathbb R}^{n \times m}$ , respectively. Let $T$ be a positive constant. We consider the following $n$ -dimensional SDDE:

$\begin{align} dX(t)& = f(X(t), \ldots, X(\delta_{\ell}(t))) dt + g(X(t), \ldots, X(\delta_{\ell}(t)))dB(t), \quad 0 \leq t \leq T, \\ X(t)& = \xi(t), \quad -\tau \leq t\leq 0. \end{align}$

(2.5)

For the functions $f$ and $g$ , we assume the following:

(H) For any $x_0, \ldots, x_{\ell}, \overline{x}_0, \ldots, \overline{x}_{\ell} \in \mathbb{R}^n$ , there exists $K_2 > 0$ such that

$\begin{align} &\left| f(x_0, \ldots, x_{\ell})-f(\overline{x}_0, \ldots, \overline{x}_{\ell}) \right|^2 \vee |g(x_0, \ldots, x_{\ell})-g(\overline{x}_0, \ldots, \overline{x}_{\ell})|^2 \\ \leq & K_2(|x_0-\overline{x}_0|^2+\cdots +|x_{\ell}-\overline{x}_{\ell}|^2), \end{align}$

(2.6)

where $a \vee b: = \max\{a, b\}$ .

We remark that (H) implies that

$\begin{align} |f(x_{0}, \ldots, x_{\ell})|^2 \vee|g(x_{0}, \ldots, x_{\ell})|^2 \leq & K(1+|x_{0}|^2 + \cdots +|x_{\ell}|^2), \end{align}$

(2.7)

where

$\begin{align} K: = 2(K_{2}\vee |f(0, \ldots, 0)|^2 \vee |g(0, \ldots, 0)|^2). \end{align}$

(2.8)

2.2. Euler–Maruyama approximation

We next explain the Euler–Maruyama approximation scheme for the SDDE (2.5). We set a time step $\varDelta$ with an integer $N$ such that

$\begin{align} \varDelta: = \tau/N \leq \frac{1}{\rho+1}. \end{align}$

(2.9)

Using the time step, we provide a discrete approximate solution of the SDDE (2.5) by

$\begin{align} \overline{y}((k+1)\varDelta) = &\overline{y}(k\varDelta) +f\left(\overline{y}(k\varDelta), \ldots, \overline{y}(I_{k\varDelta}^{(\ell)}\varDelta) \right)\varDelta +g\left( \overline{y}(k\varDelta), \ldots, \overline{y}(I_{k\varDelta}^{(\ell)}\varDelta) \right)\varDelta B_{k}, \\ & \;\; \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad k = 0, 1, \ldots, N-1, \\ \overline{y}(t) = &\xi(t), \quad -\tau \leq t\leq 0, \end{align}$

(2.10)

where $\varDelta B_{k} = B((k+1)\varDelta)-B(k \varDelta)$ and $I_{k\varDelta}^{(i)}$ is the integer part of $\delta_i(k \varDelta)/\varDelta$ for $i = 1, 2, \ldots, \ell$ .

We consider a continuous approximate solution. Let $1_S(t)$ denote the indicator function of a subset of time interval $S$ . We set

$\begin{align} z_{0}(t): = &\sum\limits_{k = 0}^{\infty}1_{[k\varDelta, (k+1)\varDelta)}(t) \overline{y}(k\varDelta), \\ z_{i}(t): = &\sum\limits_{k = 0}^{\infty}1_{[k\varDelta, (k+1)\varDelta)}(t) \overline{y}(I_{k \varDelta}^{(i)} \varDelta), \quad i = 1, \ldots, \ell, \end{align}$

(2.11)

and define a continuous Euler–Maruyama approximate solution:

$\begin{align*} y(t) = \begin{cases} &\xi(t), \quad -\tau \leq t\leq 0, \\ &\xi(0)+ {\int_{0}^{t}} f(z_{0}(s), \ldots, z_{\ell}(s))ds + {\int_{0}^{t}} g(z_{0}(s), \ldots, z_{\ell}(s))dB(s), \ 0\leq t\leq T. \end{cases} \end{align*}$

We remark that for each $k$ , the discrete solution $\bar{y}(k \varDelta)$ and the continuous solution $y(k \varDelta)$ are the same. Then, we obtain the rate of convergence in the $L^p$ -mean as follows.

Theorem 2.1. For the SDE with multiple delays (2.5), we assume $($ D $)$ , $($ P $)$ and $($ H $)$ . Then, for $p \geq 2$ in $($ P $)$ there exist constants $C_1$ and $C_2$ such that

$\begin{align} &E \left[ \sup\limits_{0\leq t \leq T}|X(t)-y(t)|^p \right] \\ \leq& 4^{p-1} K_2^{p/2} T^p (1+c_{p}) (\ell+2)^{p/2-1} (C_1+ \ell C_2)\ \varDelta^{\gamma} \cdot \exp \left[ 4^{p-1} K_2^{p/2} T^p (1+c_{p}) (\ell+2)^{p/2-1} (\ell+1) \right], \end{align}$

(2.12)

where $c_p = \frac{p^{p(p+1)/2}}{2^{p/2} (p-1)^{p(p-1)/2}}$ .

Remark 2.1. We remark that rates of convergence of approximate solutions of SDEs in the $L^p$ -mean are given by the strong order to the power $p/2$ $($ e.g. Theorem 1 in ^[11]). For SDDEs, the rates of convergence are influenced by not only $p$ but also the Hölder continuous of initial data $\gamma$ .

We provide the proof of Theorem 2.1 in the final section.

3. Confidence interval estimations for approximate solutions

In this section, we consider confidence interval estimations for approximate solutions. Theorem 2.1 implies an error estimation for Euler-Maruyama approximate solutions of diffusion processes governed by SDEs with multiple delays. The inequality (2.12) tells us

$\begin{align*} \lim\limits_{\varDelta \to 0} E \left[ \sup\limits_{0\leq t \leq T}|X(t)-y(t)|^2 \right] = 0. \end{align*}$

The definition of time step (2.9) and Chebyshev's inequality imply that for any $\epsilon > 0$ ,

$\begin{align*} P\left\{ \sup\limits_{0 \leq t \leq T} |X(t)-y(t)| \leq \epsilon \right\} \geq& 1- E\left[ {\sup\limits_{0 \leq t \leq T}} |X(t)-y(t)|^2 \right]/\epsilon^2 \\ \geq& 1-O(N^{-1})/\epsilon^2 \quad \mbox{as}\ N\ \to \infty. \end{align*}$

This inequality provides only the order of the hazard rate; hence, more information is required to obtain a confidence interval estimation. In this section, we present refined estimations for a general case and a special case.

3.1. General case

Using the inequality (2.12) in Theorem 2.1 with $p = 2$ , we obtain the following confidence interval estimation.

Theorem 3.1. For the SDE with multiple delays (2.5), we assume $($ D $)$ , $($ P $)$ and $($ H $)$ . Then, for any $\epsilon > 0$

$\begin{align} P \left[\sup\limits_{0 \leq t \leq T} |X(t)-y(t)|\leq \epsilon \right] \geq & 1- \frac{20}{\epsilon^2} K_2 T^2(C_1+ \ell C_2) e^{20 K_2 T^2(\ell +1)} \varDelta^{\gamma}. \end{align}$

(3.1)

Proof. Theorem 2.1 and Chebyshev's inequality imply that

$\begin{align*} P\left\{ \sup\limits_{0 \leq t \leq T} |X(t)-y(t)| > \epsilon \right\} < & \frac{1}{\epsilon^2} E \left[ \sup\limits_{0 \leq t \leq T} |X(t)-y(t)|^2 \right] \\ < & \frac{20}{\epsilon^2} K_2 T^2(C_1+ \ell C_2) e^{20 K_2 T^2 (\ell+1)} \varDelta^{\gamma}, \end{align*}$

which proves Theorem 3.1.

3.2. Special case

The expressions of constants $C_1$ and $C_2$ in Theorems 2.1 are complicated, and they are given in (5.5) and (5.9), respectively (see also (5.7) in Lemma 5.3). For SDDEs which have no drift terms, the expressions become simpler. We next consider the special case and provide confidence interval estimations.

For the case $f \equiv 0$ in (2.5), the SDDE is given by

$\begin{align} d \widetilde{X}(t)& = g \left( \widetilde{X}(t), \ldots, \widetilde{X}(\delta_{\ell}(t)) \right) dB(t), \quad 0 \leq t \leq T, \\ \widetilde{X}(t)& = \xi(t), \quad -\tau \leq t\leq 0. \end{align}$

(3.2)

$\widetilde{y}(t)$ denotes a discreate Euler–Maruyama approximate solution of $\widetilde{X}(t)$ . Then, we obtain the following confidence interval estimation.

Corollary 3.1. For the SDE with multiple delays (3.2), we assume $($ D $)$ , $($ P $)$ and $($ H $)$ . Then, there exist constants $\widetilde{C}_1$ and $\widetilde{C}_2$ such that, for any $\epsilon > 0$

$\begin{align} &P \left[\sup\limits_{0 \leq t \leq T} \left| \widetilde{X}(t)-\widetilde{y}(t) \right| \leq \epsilon \right] \geq 1- \frac{8}{\epsilon^2} K_2 T(\widetilde{C}_1+ \ell \widetilde{C}_2) e^{8 K_2 T (\ell +1)} \varDelta^{\gamma}, \end{align}$

(3.3)

where

$\begin{align*} \widetilde{C}_1 = K \{ 1+ \widetilde{C}_3 (\ell+1) \}, \ \widetilde{C}_3 = 2(K_0+KT)e^{2 KT (\ell+1)}, \end{align*}$

and $\widetilde{C}_2$ is a constant which is given as follows:

for delay functions $\delta_i(t), i = 1, \ldots, \ell$ ,

$\begin{align} \widetilde{C}_2^{i} = \left\{ \begin{array}{ll} \widetilde{C}_1 (\rho+1), \qquad &0 \leq I_{k\varDelta}^{(i)} \varDelta \leq \delta_{i}(t), \\[.1cm] \widetilde{C}_1 \rho, \qquad &0 \leq \delta_{i}(t) \leq I_{k\varDelta}^{(i)} \varDelta, \\[.1cm] 2\widetilde{C}_1 (\rho+1)+2K_{1}(\rho+1)^{\gamma}, \qquad &I_{k\varDelta}^{(i)} \varDelta \leq 0 \leq \delta_{i}(t), \\[.1cm] 2\widetilde{C}_1 \rho+2K_{1}\rho^{\gamma}, \qquad &\delta_{i}(t) \leq 0 \leq I_{k\varDelta}^{(i)}\varDelta, \\[.1cm] K_{1}(\rho+1)^{\gamma}, \qquad &I_{k\varDelta}^{(i)} \varDelta \leq \delta_{i}(t) \leq 0\ \mathit{\text{or}}\ \delta_{i}(t) \leq I_{k\varDelta}^{(i)} \varDelta \leq 0, \end{array} \right. \end{align}$

(3.4)

and we set

$\begin{align} \widetilde{C}_2: = \max\limits_{i = 1, \ldots, \ell} \widetilde{C}_2^i. \end{align}$

(3.5)

4. Numerical examples

4.1. Multi-dimensional SDDE

Example 4.1. Recall that $B(t) = {}^t(B_1(t), \ldots, B_m(t))$ is an $m$ -dimensional standard Brownian motion. Given an $n$ -dimensional vector function $f = (f_1, \ldots, f_n)$ and an $n \times m$ -matrix function $g = (g_{ij})$ , we consider the following $n$ -dimensional SDDE:

$\begin{align} dX_l(t) = f_l(X(t), \ldots, X(\delta_{\ell}(t))) dt + \sum\limits_{j = 1}^m g_{lj}(X(t), \ldots, X(\delta_{\ell}(t))) dB_j(t), \ 0 \leq t \leq T, \end{align}$

(4.1)

which is the $l$ -th component of the $n$ -dimensional SDDE. We assume that Lipschitz's constant in $(2.2)$ is given by $\rho = 10^{-1}$ . To apply Theorem 3.1, we give the setting as follows:

Finish time: $T = 1;$

Time step: $\varDelta = 10^{-4};$

Uniform boundedness of initial data: $K_0 = 10^{-1};$

Hölder continuity of initial data: $\gamma = 1, K_1 = 10^{-1};$

The constant $K_2$ in (2.6): $K_2 = 10^{-2(n-1)}$ .

In the case of $\ell = n = 2$ , we set ${C}_2$ as given in (5.9) $($ see also (5.7) in Lemma 5.3 $)$ . Then, Theorem 3.1 implies that

$\begin{align*} P \left\{ \sup\limits_{0 \leq t \leq 1} \left| X(t)-y(t) \right| \leq 1.94 \times 10^{-2} \right\} \geq 0.9. \end{align*}$

Under the same setting as listed above, we present $\epsilon$ 's in (3.1) for other cases of $\ell$ and $n$ with the confidence levels 0.9 and 0.95 in Tables 1 and 2, respectively.

Table 1. Numerical results for

$n$ and

$\ell$ on Example 4.1 with the confidence level

$0.9$ .

$n \backslash \ell$	1	2	3	4	5	10
2	$1.068 \times 10^{-2}$	$1.944 \times 10^{-2}$	$3.028 \times 10^{-2}$	$4.404 \times 10^{-2}$	$6.156 \times 10^{-2}$	$2.470 \times 10^{-1}$
3	$8.760 \times 10^{-4}$	$1.444 \times 10^{-3}$	$2.038 \times 10^{-3}$	$2.684 \times 10^{-3}$	$3.399 \times 10^{-3}$	$8.314 \times 10^{-3}$
4	$8.742 \times 10^{-5}$	$1.440 \times 10^{-4}$	$2.030 \times 10^{-4}$	$2.671 \times 10^{-4}$	$3.379 \times 10^{-4}$	$8.224 \times 10^{-4}$

| Show Table

DownLoad: CSV

Table 2. Numerical results for

$n$ and

$\ell$ on Example 4.1 with the confidence level

$0.95$ .

$n \backslash \ell$	1	2	3	4	5	10
2	$1.151 \times 10^{-2}$	$2.749 \times 10^{-2}$	$4.282 \times 10^{-2}$	$6.228 \times 10^{-2}$	$8.706 \times 10^{-2}$	$3.494 \times 10^{-1}$
3	$1.239 \times 10^{-3}$	$2.043 \times 10^{-3}$	$2.882 \times 10^{-3}$	$3.797 \times 10^{-3}$	$4.807 \times 10^{-3}$	$1.176 \times 10^{-2}$
4	$1.236 \times 10^{-4}$	$2.037 \times 10^{-4}$	$2.870 \times 10^{-4}$	$3.778 \times 10^{-4}$	$4.778 \times 10^{-4}$	$1.163 \times 10^{-3}$

| Show Table

DownLoad: CSV

4.2. One-dimensional SDDE

We consider the following SDE with piecewise constant arguments (Example 1 in ^[19]):

$\begin{align*} dX(t)& = \{ X(t)+X([t]) +X([t-1])\}dt + \{X(t)+X([t-1])\} dB(t), \ 0 \leq t \leq 2, \\ X(t)& = 1, \ -1 \leq t \leq 0. \end{align*}$

Since the example above has piecewise constant time delays, the explicit solution is given by

$\begin{align*} X(t) = \left\{ \begin{array}{l} { \exp \left\{ \frac{t}{2} + B(t) \right\} \left( \xi(0)+\int_0^t e^{-\frac{s}{2}-B(s)}ds +\int_0^t e^{-\frac{s}{2}-B(s)}dB(s) \right)}, \\ \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad t \in [0, 1], \\ { \exp\left\{ \frac{t-1}{2} + B(t)-B(1) \right\} \left( X(1)+ X(1) \int_1^t e^{-\frac{s-1}{2}-\{B(s)-B(1)\}}\}ds +\int_0^t e^{-\frac{s-1}{2}-\{B(s)-B(1)\}}dB(s) \right)}, \\ \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad t \in [1, 2]. \end{array} \right. \end{align*}$

In the case of delay functions, the representations of explicit solutions of SDDEs by the stochastic integral are complicated. In such a case, we consider approximate solutions for SDDEs.

Example 4.2. We consider the following one-dimensional SDDE with two-time delay functions.

$\begin{align} dX(t)& = \{ X(t)+X(\delta_1(t)) +X(\delta_2(t))\}dt + \{ X(t)+X(\delta_1(t))+X(\delta_2(t))\} dB(t), \end{align}$

(4.2)

where Lipschitz's constant in (2.2) is given by $\rho = 10^{-1}$ . To apply Theorem 3.1, we give the setting as follows:

Finish time: $T = 2;$

Time step: $\varDelta = 10^{-4};$

Uniform boundedness of initial data: $K_0 = 10^{-1};$

Hölder continuity of initial data: $\gamma = 1, K_1 = 10^{-1};$

The constant $K_2$ in (2.6): $K_2 = 10^{-2}$ .

We set ${C}_2$ in the same manner as that in Example 4.1. Then, Theorem 3.1 implies that

$\begin{align*} P \left\{ \sup\limits_{0 \leq t \leq 1} \left| X(t)-y(t) \right| \leq 0.15 \right\} \geq 0.9. \end{align*}$

4.3. Special case

Example 4.3. We consider the SDDE (3.2). To apply Corollary 3.1, we give the setting as follows:

Finish time: $T = 1;$

Time step: $\varDelta = 10^{-4};$

Lipschitz constant of the time delay: $\rho = 10^{-1};$

Uniform boundedness of initial data: $K_0 = 10^{-1};$

Hölder continuity of initial data: $\gamma = 1, K_1 = 10^{-1};$

The constant $K_2$ in (2.6): $K_2 = 10^{-2(n-1)}$ .

Under the conditions above, $\widetilde{C}_2$ in (3.5) is determined. In the case of $\ell = n = 2$ , we obtain that

$P \left\{ \sup\limits_{0 \leq t \leq 1} \left| \widetilde{X}(t)-\widetilde{y}(t) \right| \leq 8.04 \times 10^{-3} \right\} \geq 0.9.$

Under the same setting as listed above, we present $\epsilon$ 's in (3.3) for other cases of $\ell$ and $n$ with the confidence levels 0.9 and 0.95 in Tables 3 and 4, respectively.

Table 3. Numerical results for

$n$ and

$\ell$ on Example 4.3 with the confidence level

$0.9$ .

$n \backslash \ell$	1	2	3	4	5	10
2	$5.458 \times 10^{-3}$	$8.040 \times 10^{-3}$	$1.041 \times 10^{-2}$	$1.276 \times 10^{-2}$	$1.512 \times 10^{-2}$	$2.961 \times 10^{-2}$
3	$4.797 \times 10^{-4}$	$6.935 \times 10^{-4}$	$8.691 \times 10^{-4}$	$1.027 \times 10^{-3}$	$1.177 \times 10^{-3}$	$1.893 \times 10^{-3}$
4	$4.791 \times 10^{-5}$	$6.925 \times 10^{-5}$	$8.675 \times 10^{-5}$	$1.025 \times 10^{-4}$	$1.174 \times 10^{-4}$	$1.886 \times 10^{-4}$

| Show Table

DownLoad: CSV

Table 4. Numerical results for

$n$ and

$\ell$ on Example 4.3 with the confidence level

$0.95$ .

$n \backslash \ell$	1	2	3	4	5	10
2	$7.718 \times 10^{-3}$	$1.137 \times 10^{-2}$	$1.472 \times 10^{-2}$	$1.804 \times 10^{-2}$	$2.145 \times 10^{-2}$	$4.187 \times 10^{-2}$
3	$6.784 \times 10^{-4}$	$9.808 \times 10^{-4}$	$1.229 \times 10^{-3}$	$1.453 \times 10^{-3}$	$1.664 \times 10^{-3}$	$2.678 \times 10^{-3}$
4	$6.775 \times 10^{-5}$	$9.794 \times 10^{-5}$	$1.227 \times 10^{-4}$	$1.450 \times 10^{-4}$	$1.660 \times 10^{-4}$	$2.667 \times 10^{-4}$

| Show Table

DownLoad: CSV

5. Proofs

5.1. Proof of Theorem 2.1.

Lemma 5.1. Under the assumptions (P) (i) and (H),

$\begin{align} \sup\limits_{t \in [-\tau, T]}E[|y(t)|^p] \leq 3^{p-1} \left\{ K_0 + C_3 T \right\} \cdot \exp\{ 3^{p-1} C_3(\ell+1)T \}, \end{align}$

(5.1)

where

$\begin{align} C_3 = K^{p/2}(\ell+2)^{p/2-1} \left( T^{p-1}+T^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right). \end{align}$

(5.2)

Proof. (ⅰ) For the case $-\tau \leq t\leq 0$ ,

$\begin{align*} E[|y(t)|^p] \leq \sup\limits_{-\tau \leq t\leq 0}E[|\xi(t)|^p] = K_0. \end{align*}$

(ⅱ) For the case $0 \leq t\leq T$ ,

$\begin{align*} &E[|y(t)|^p]\\ = &E \left[ \left| \xi(0)+\int_{0}^{t}f(z_{0}(s), \ldots, z_{\ell}(s))ds +\int_{0}^{t}g(z_{0}(s), \ldots, z_{\ell}(s))dB(s) \right|^p \right]\\ \leq& 3^{p-1} \left\{ E[|\xi(0)|^p] +E \left[ \left| \int_{0}^{t}f(z_{0}(s), \ldots, z_{\ell}(s))d s \right|^p \right] +E \left[ \left| \int_{0}^{t}g(z_{0}(s), \ldots, z_{\ell}(s))dB(s) \right|^p \right] \right\} \\ = :& 3^{p-1}(I_1+I_2+I_3). \end{align*}$

Using Hölder's inequality and the inequality (2.7), we obtain

$\begin{align*} I_2 \leq& t^{p-1} \cdot E \left[ \int_0^t |f(z_{0}(s), \ldots, z_{\ell}(s))|^p ds \right] \\ \leq & K^{p/2} t^{p-1} E \left[ \int_0^t \{1+|z_0(s)|^2+ \cdots + |z_{\ell}(s)|^2 \}^{p/2} ds \right] \\ \leq & K^{p/2} (\ell+2)^{p/2-1} t^{p-1} E \left[ \int_0^t \{1+|z_0(s)|^p+ \cdots + |z_{\ell}(s)|^p \} ds \right]. \end{align*}$

Using Burkholder–Davis–Gundy's inequality, we obtain

$\begin{align*} I_3 \leq& t^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} E \left[ \int_0^t |g(z_{0}(s), \ldots, z_{\ell}(s))|^p ds \right] \\ \leq & K^{p/2} t^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} E \left[ \int_0^t \{1+|z_0(s)|^2+ \cdots + |z_{\ell}(s)|^2 \}^{p/2} ds \right] \\ \leq & K^{p/2} ({\ell}+2)^{p/2-1} t^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} E \left[ \int_0^t \{1+|z_0(s)|^p+ \cdots + |z_{\ell}(s)|^p \} ds \right]. \end{align*}$

Thus,

$\begin{align*} I_2+I_3 &\leq K^{p/2}(\ell+2)^{p/2-1} \left( t^{p-1}+ t^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right) \left\{t+(\ell+1) \int_0^t \sup\limits_{-\tau \leq u \leq s} E[|y(u)|^p] ds \right\} \\ &\leq C_3 \left\{t+(\ell+1) \int_0^t \sup\limits_{-\tau \leq u \leq s} E[|y(u)|^p] ds \right\}. \end{align*}$

Using Gronwall's lemma, we obtain

$\begin{align*} \sup\limits_{-\tau \leq t \leq T} E[|y(t)|^p] \leq 3^{p-1} \left\{ K_0 +C_3 T \right\} \cdot \exp\{3^{p-1} C_3 (\ell+1)T\}.\ \Box \end{align*}$

Following Lemma 5.1, we set

$\begin{align} C_4: = 3^{p-1} \left\{ K_0 +C_3 T \right\} \cdot \exp\{3^{p-1} C_3(\ell+1)T\}, \end{align}$

(5.3)

which does not depend on $\varDelta$ .

Lemma 5.2. Under the assumptions $($ D $)$ , $($ P $)$ $($ i $)$ and $($ H $)$ , for any $t \in [0, T]$

$\begin{align} E[|y(t)-z_{0}(t)|^p] \leq 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( \varDelta^{p-1}+\varDelta^{p/2-1} \left\{\frac{p(p-1)}{2}\right\}^{p/2} \right) \{1+C_4(\ell+1)\} \varDelta. \end{align}$

(5.4)

Proof. For a fixed $t$ , there exists $k$ such that $t \in [k \varDelta, (k+1)\varDelta)$ . We remark that $z_0(t) = \bar{y}(k \varDelta)$ . In the same manner as those for showing Lemma 5.1, we obtain

$\begin{align*} & E[|y(t)-z_0(t)|^p] = E[|y(t)-\bar{y}(k \varDelta)|^p] \\ \leq & E\left[ \left| \int_{k \varDelta}^t f(z_0(s), \ldots, z_{\ell}(s)) ds + \int_{k \varDelta}^t g(z_0(s), \ldots, z_{\ell}(s))dB(s) \right|^p \right] \\ \leq & 2^{p-1}\left\{ E \left[ \left| \int_{k\varDelta}^t f(z_0(s), \ldots, z_{\ell}(s)) ds \right|^p \right] + E \left[ \left| \int_{k\varDelta}^t g(z_0(s), \ldots, z_{\ell}(s)) dB(s) \right|^p \right] \right\} \\ \leq & 2^{p-1} K^{p/2} ({\ell}+2)^{p/2-1} \left( \varDelta^{p-1} + \varDelta^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2} \right) E\left[ \int_{k \varDelta}^t \left\{1+|z_0(s)|^p+ \cdots +|z_{\ell}(s)|^p\right\} ds \right] \\ \leq & 2^{p-1} K^{p/2} ({\ell}+2)^{p/2-1} \left( \varDelta^{p-1}+\varDelta^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1)\} \varDelta.\ \Box \end{align*}$

Following Lemma 5.2, we set

$\begin{align} C_1: = 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( 1+\left\{\frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+C_4 (\ell+1)\}, \end{align}$

(5.5)

which does not depend on $\varDelta$ either.

Lemma 5.3. Under the assumptions $($ D $)$ , $($ P $)$ and $($ H $)$ , for any $t \in [0, T]$

$\begin{align} E[|y(\delta_{i}(t))-z_{i}(t)|^p] \leq {C_2^i} \varDelta^{\gamma}, \ i = 1, \ldots, \ell, \end{align}$

(5.6)

where $C_2^i$ is given as follows:

$\begin{align} {C_2^i} = \begin{cases} &2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( (\rho+1)^{p-1} + (\rho+1)^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2}\right) \{1+C_4 (\ell+1)\}(\rho+1), \\ &\hfill 0 \leq I_{k \varDelta}^{(i)}\varDelta \leq \delta_{i}(t), \\ & 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( \rho^{p-1} + \rho^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2}\right) \{1+ C_4 (\ell+1) \} \rho, \\ &\hfill 0 \leq \delta_{i}(t) \leq I_{k \varDelta}^{(i)}\varDelta, \\ & 2^{p-1} \biggl\{ 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( (\rho+1)^{p-1} + (\rho+1)^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1) \} (\rho+1) \\ & \hfill + {K}_1(\rho+1)^{\gamma} \biggr\}, \\ &\hfill I_{k \varDelta}^{(i)}\varDelta \leq 0 \leq \delta_i(t), \\ & 2^{p-1} \left\{K_1 \rho^{\gamma} + 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( (\rho+1)^{p-1} + (\rho+1)^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1) \} \rho \right\}, \\ & \hfill \delta_{i}(t) \leq 0 \leq I_{k \varDelta}^{(i)}\varDelta, \\ &K_1 (\rho+1)^{\gamma}, \hfill I_{k \varDelta}^{(i)}\varDelta \leq \delta_{i}(t) \leq 0 \ {{or}}\ \delta_{i}(t) \leq I_{k \varDelta}^{(i)}\varDelta \leq 0. \end{cases} \end{align}$

(5.7)

Proof. (Ⅰ) Case in which $0\leq I_{k \varDelta}^{(i)}\varDelta \leq \delta_{i}(t)$ :

For a fixed $t \in [0, T]$ , there exists $k$ such that $t \in [k\varDelta, (k+1)\varDelta)$ . Then,

$\begin{align*} y(\delta_{i}(t))-z_{i}(t) = y(\delta_{i}(t))-y(I_{k \varDelta}^{(i)}\varDelta). \end{align*}$

As $\delta_{i}(t)-I_{k \varDelta}^{(i)}\varDelta \leq (\rho+1)\varDelta$ ,

$\begin{align*} &E[|y(\delta_{i}(t))-z_{i}(t)|^p]\\ \leq& 2^{p-1} \left\{ E \left[ \left| \int_{I_{k \varDelta}^{(i)}\varDelta}^{\delta_{i}(t)} f(z_{0}(s), \ldots, z_{\ell}(s))ds \right|^p \right] + E \left[ \left| \int_{I_{k \varDelta}^{(i)}\varDelta}^{\delta_{i}(t)} g(z_{0}(s), \ldots, z_{\ell}(s))dB(s) \right|^p \right] \right\} \\ \leq& 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( \{(\rho+1) \varDelta \}^{p-1} + \{(\rho+1) \varDelta\}^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+C_4 (\ell+1) \}(\rho+1) \varDelta. \end{align*}$

This inequality and $\varDelta < 1$ imply a constant $C_2^i$ for this case.

(Ⅱ) Case in which $0\leq \delta_i(t) \leq I_{k \varDelta}^{(i)}\varDelta$ :

As $I_{k \varDelta}^{(i)}\varDelta - \delta_i(t) \leq \delta_i(k \varDelta)-\delta_i(t) \leq \rho \varDelta$ ,

$\begin{align*} &E[|y(\delta_{i}(t))-z_{i}(t)|^p]\\ \leq & 2^{p-1} \left\{ E \left[ \left| \int_{\delta_i(t)}^{I_{k \varDelta}^{(i)}\varDelta} f(z_{0}(s), \ldots, z_{\ell}(s))ds \right|^p \right] + E \left[ \left| \int_{\delta_i(t)}^{I_{k \varDelta}^{(i)}\varDelta} g(z_{0}(s), \ldots, z_{\ell}(s))dB(s) \right|^p \right] \right\} \\ \leq & 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( (\rho \varDelta)^{p-1} + (\rho \varDelta)^{p/2-1} \left\{\frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1) \} \rho \varDelta. \end{align*}$

(Ⅲ) Case in which $I_{k \varDelta}^{(i)}\varDelta \leq 0 \leq \delta_i(t)$ :

As $\delta_i(t) \leq \delta_i(t)-I_{k \varDelta}^{(i)}\varDelta \leq (\rho+1) \varDelta$ , Lemma 5.2, and (P) (ⅱ) imply that

$\begin{align*} &E[|y(\delta_{i}(t))-z_{i}(t)|^p]\\ \leq & 2^{p-1} \left\{ E \left[ |y(\delta_i(t)) - \xi(0)|^p \right] +E[|\xi(0)-\xi(I_{k \varDelta}^{(i)}\varDelta)|^p ] \right\}\\ \leq & 2^{p-1} \Biggl\{ 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( \delta_i(t)^{p-1} + \delta_i(t)^{p/2-1} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1) \} \delta_i(t) \\ & \quad + K_1(-I_{k \varDelta}^{(i)}\varDelta)^{\gamma} \Biggr\} \\ \leq & 2^{p-1} \Biggl\{ 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( (\rho+1)^{p-1} \varDelta^{p-\gamma} + (\rho+1)^{p/2-1} \varDelta^{p/2 - \gamma} \left\{ \frac{p(p-1)}{2} \right\}^{p/2} \right) \{1+ C_4 (\ell+1) \} (\rho+1) \\ & \quad + {K}_1(\rho+1)^{\gamma} \Biggr\} \varDelta^{\gamma}. \end{align*}$

(Ⅳ) Case in which $\delta_i(t) \leq 0 \leq I_{k \varDelta}^{(i)}\varDelta$ :

As $I_{k \varDelta}^{(i)}\varDelta \leq I_{k \varDelta}^{(i)}\varDelta - \delta_i(t) \leq \delta_i(k \varDelta) - \delta_i(t)\leq \rho \varDelta$ , in the same manner as that in case (Ⅲ) we obtain

$\begin{align*} &E[|y(\delta_{i}(t))-z_{i}(t)|^p] \\ \leq & 2^{p-1} \left\{ E \left[ |y(\delta_i(t)) - \xi(0)|^p \right] +E[|\xi(0)-\xi(I_{k \varDelta}^{(i)}\varDelta)|^p ] \right\} \\ \leq & 2^{p-1} \Biggl\{K_1 \rho^{\gamma} + 2^{p-1} K^{p/2} (\ell+2)^{p/2-1} \left( {(\rho+1)^{p-1} \varDelta^{p-\gamma} + (\rho+1)^{p/2-1} \varDelta^{p/2 - \gamma} \left\{ \frac{p(p-1)}{2} \right\}^{p/2}} \right) \nonumber\\ & \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \{1+ C_4 (\ell+1) \} \rho \Biggr\} \varDelta^{\gamma}. \end{align*}$

(Ⅴ) Cases in which $I_{k \varDelta}^{(i)}\varDelta \leq \delta_i(t) \leq 0$ or $\delta_i(t) \leq I_{k \varDelta}^{(i)}\varDelta \leq 0$ :

As $|\delta_i(t) - I_{k \varDelta}^{(i)}\varDelta | \leq (\rho+1) \varDelta$ ,

$\begin{align} E[|y(\delta_{i}(t))-z_{i}(t)|^p] \leq & E \left[ \left| \xi(\delta_i(t)) - \xi(I_{k \varDelta}^{(i)}\varDelta) \right|^p \right] \\ \leq & K_1 \left| \delta_i(t) - I_{k \varDelta}^{(i)}\varDelta \right|^{\gamma} \\ \leq & K_1 (\rho+1)^{\gamma} \varDelta^{\gamma}.\ \Box \end{align}$

(5.8)

Following Lemma 5.3, we set

$\begin{align} C_2: = \max\limits_{i = 1, \ldots, \ell} C_2^i. \end{align}$

(5.9)

Proof of Theorem 2.1. For any $t_{1}\leq T$ ,

$\begin{align} &E[\sup\limits_{0\leq t \leq t_{1}}|X(t)-y(t)|^p] \\ = &E \left[ \sup\limits_{0\leq t \leq t_{1}} \left| \int_{0}^{t} \left\{ f \left( X(s), \ldots, X(\delta_{\ell}(s)) \right) - f \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\} ds \right.\right.\\ & \quad + \left. \left. \int_{0}^{t} \left \{ g \left( X(s), \ldots, X(\delta_{\ell}(s)) \right) -g \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\} dB(s) \right|^p \right] \\ \leq &2^{p-1} E \left[ \sup\limits_{0\leq t \leq t_{1}} \left| \int_{0}^{t} \left\{ f \left( X(s), \ldots, X(\delta_{l}(s)) \right) - f \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\} ds \right|^p \right]\\ & \quad +2^{p-1} E \left[ \sup\limits_{0\leq t \leq t_{1}} \left| \int_{0}^{t} \left\{ g \left( X(s), \ldots, X(\delta_{\ell}(s)) \right) -g \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\} dB(s) \right|^p \right]\\ = : & 2^{p-1} (I_4+I_5). \end{align}$

(5.10)

Using Hölder's inequality, for any $t_1 \leq T$ we have

$\begin{align} I_4 \leq& T^{p-1} E \left[ \left| \int_{0}^{t_{1}} \left\{ f \left( X(s), \ldots, X(\delta_{\ell}(s)) \right) -f \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\}^2 ds \right|^{p/2} \right]. \end{align}$

(5.11)

Using Burkholder–Davis–Gundy's inequality, for any $t_i \leq T$ we have

$\begin{align} I_5 \leq& T^{p-1} c_p E \left[ \left| \int_0^{t_1} \left\{ g \left( X(s), \ldots, X(\delta_{\ell}(s)) \right) -g \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right\}^2 ds \right|^{p/2} \right], \end{align}$

(5.12)

where $c_p = \frac{p^{p(p+1)/2}}{2^{p/2} (p-1)^{p(p-1)/2}}$ . The inequalities (5.11) and (5.12), and the assumption (H) imply that

$\begin{align} & (\mbox{the right-hand side of}\ (5.10)) \\ \leq & 2^{p-1} K_2^{p/2}T^{p-1} (1+c_p)(\ell+2)^{p/2-1} \cdot E \left[ \int_{0}^{t_{1}} \left\{ |X(s)-z_{0}(s)|^p + \sum\limits_{i = 1}^{\ell}|X(\delta_{i}(s))-z_{i}(s)|^p \right\} ds \right] \\ \leq & 2^{2(p-1)} K_2^{p/2}T^{p-1} (1+c_p)(\ell+2)^{p/2-1} \\ & \quad \quad\quad\quad\cdot \left\{ E \left[ \int_{0}^{t_{1}} \left\{ |X(s)-y(s)|^p + \sum\limits_{i = 1}^{\ell}|X(\delta_{i}(s))-y(\delta_i(s))|^p \right\} ds \right] \right. \\ & \quad \quad\quad\quad\quad\quad\quad\quad\quad\quad \left. + E\left[ \int_{0}^{t_{1}} \left\{ |y(s)-z_{0}(s)|^p +\sum\limits_{i = 1}^{\ell} |y(\delta_i(s))-z_i(s)|^p \right\} ds \right] \right\} \\ \leq &4^{p-1} K_2^{p/2}T^{p-1} (1+c_p)(\ell+2)^{p/2-1} \left\{(\ell+1)\int_{0}^{t_{1}} E \left[ \sup\limits_{0 \leq r\leq s}|X(r)-y(r)|^p \right] ds \right. \\ & \quad \quad\quad\quad\quad\quad\quad\quad\quad \left. + T \left( E \left[ \sup\limits_{0 \leq s\leq T}|y(s)-z_{0}(s)|^p \right] +\sum\limits_{i = 1}^{\ell} E \left[ \sup\limits_{0 \leq s \leq T} |y(\delta_i(s))-z_i(s)|^p \right] \right) \right\}. \end{align}$

(5.13)

Lemma 5.2, Lemma 5.3 and Gronwall's lemma imply that

$\begin{align*} & (\mbox{the right-hand side of}\ (5.13)) \nonumber \\ \leq& 4^{p-1} K_2^{p/2} T^{p-1} (1+c_p)(\ell+2)^{p/2-1} \left\{ (\ell+1) \int_{0}^{t_{1}} E \left[ \sup\limits_{0 \leq r\leq s}|X(r)-y(r)|^2 \right] ds + T(C_1 \varDelta + \ell C_2 \varDelta^{\gamma}) \right\} \nonumber \\ \leq& 4^{p-1} K_2^{p/2} T^p (1+c_p)(\ell+2)^{p/2-1} (C_1 \varDelta + \ell C_2 \varDelta^{\gamma}) \cdot \exp \left[4^{p-1} K_2^{p/2} T^p (1+c_p) (\ell+2)^{p/2-1} (\ell+1) \right] \nonumber \\ \leq & 4^{p-1} K_2^{p/2} T^p (1+c_{p}) (\ell+2)^{p/2-1} (C_1+ \ell C_2) \cdot \exp \left[ 4^{p-1} K_2^{p/2} T^p (1+c_{p})(\ell+2)^{p/2-1} (\ell+1) \right] \varDelta^{\gamma}.\ \Box \end{align*}$

5.2. Proof of Corollary 3.1.

We prove Corollary 3.1 in the following steps:

1st step (corresponding to Lemma 5.1)

For the case $-\tau \leq t \leq 0$ , we obtain that $E[|\widetilde{y}(t)|^2] \leq K_0$ .

For the case $0 \leq t \leq T$ , we obtain that

$\begin{align*} E[|\widetilde{y}(t)|^2] = & E \left[ \left| \xi(0)+\int_{0}^{t}g(z_{0}(s), z_{1}(s), \ldots, z_{l}(s))dB(s) \right|^2 \right]\\ \leq& 2 \left\{ E[|\xi(0)|^2] +E \left[ \int_{0}^{t}|g(z_{0}(s), z_{1}(s), \ldots, z_{l}(s)|^2 dB(s) \right] \right\}. \end{align*}$

This inequality and Gronwall's lemma imply that

$\begin{align} \sup\limits_{-\tau \leq t \leq T} E[|\widetilde{y}(t)|^2] \leq 2 (K_0 + KT) e^{2K T(\ell+1)} = \widetilde{C}_3. \end{align}$

(5.14)

2nd step (corresponding to Lemma 5.2)

We consider the approximate solution of SDDE (2.10) with $f \equiv 0$ and use the same notation $z_i(t), i = 0, 1, \ldots, \ell$ in (2.11) in the following step. We obtain

$\begin{align} E[|\widetilde{y}(t)-z_{0}(t)|^2] \leq & E \left[ \left| \int_{k\varDelta}^{t}g(z_{0}(s), z_{1}(s), \ldots, z_{l}(s))dB(s) \right|^2 \right] \nonumber \\ \leq & K E\left[ \int_{k \varDelta}^t \left\{1+|z_0(s)|^p+ \cdots +|z_{\ell}(s)|^p\right\} ds \right] \\ \leq& K \left\{ 1+ \widetilde{C}_3 (\ell +1) \right\} \varDelta = \widetilde{C}_1 \varDelta. \end{align}$

(5.15)

3rd step (corresponding to Lemma 5.3)

In the fifth case in (3.4), we can show that the constant $\widetilde{C}_2^i$ coincides with $C_2^i$ by using the inequality (5.8) with $p = 2$ .

We next provide the proof for the third case in (3.4). As $\delta_{i}(t) \leq \delta_{i}(t)-I_{k \varDelta}^{(i)} \varDelta \leq (\rho+1) \varDelta$ , we have

$\begin{align} E \left[ |\widetilde{y}(\delta_{i}(t))-z_{i}(t)|^2 \right] \leq & 2\left\{ E \left[|\widetilde{y}(\delta_{i}(t))-\xi(0)|^2 \right] +E \left[ \left|\xi(0) -\xi \left( I_{k \varDelta}^{(i)} \varDelta \right) \right|^2\right] \right\} \\ = : &2\{I_6+I_7\}. \end{align}$

(5.16)

Using the estimation (5.15) with $t = 0$ , we have

$\begin{align} I_6 \leq \widetilde{C}_1 (\rho+1) \varDelta. \end{align}$

(5.17)

Using the estimate (5.8), we have

$\begin{align} I_7 \leq K_1 (\rho+1)^{\gamma} \varDelta^{\gamma}. \end{align}$

(5.18)

Then, (5.17) and (5.18) imply that

$\begin{align*} (\mbox{Right-hand side of}\ (5.16)) \leq 2 \left\{ \widetilde{C}_1 (\rho+1) +K_1(\rho+1)^{\gamma} \right\} \varDelta^{\gamma}. \end{align*}$

For the other three cases in (3.4), proofs are given in the same manner as those for showing the cases (Ⅰ), (Ⅱ) and (Ⅳ) in the proof of Lemma 5.3.

We set $\widetilde{C}_2$ as (3.5). Then, for any $t \in [0, T]$ we have

$\begin{align} E[|\widetilde{y}(\delta_i(t)) -z_i(t)|^2] \leq \widetilde{C}_2 \varDelta^{\gamma}. \end{align}$

(5.19)

4th step

In the case of $f \equiv 0$ , we obtain that

$\begin{align} &E \left[ \sup\limits_{0\leq t \leq t_{1}} \left| \widetilde{X}(t)-\widetilde{y}(t) \right|^2 \right] \\ \leq & 4E \left[ \int_{0}^{t_{1}} \left| g \left( \widetilde{X}(s), \ldots, \widetilde{X}(\delta_{\ell}(s)) \right) -g \left( z_{0}(s), \ldots, z_{\ell}(s) \right) \right|^2ds \right] \\ \leq & 4K_{2}E \left[ \int_{0}^{t_{1}} \left( \left| \widetilde{X}(s)-z_{0}(s) \right|^2 + \sum\limits_{i = 1}^{\ell} \left| \widetilde{X}(\delta_{i}(s))-z_{i}(s) \right|^2 \right) ds \right] \\ \leq & 8K_{2}E \left[ \int_{0}^{t_{1}} \left( \left| \widetilde{X}(s)-\widetilde{y}(s) \right|^2 + \sum\limits_{i = 1}^{\ell} \left| \widetilde{X}(\delta_{i}(s))-\widetilde{y}(\delta_{i}(t)) \right|^2 \right) ds \right] \\ & \quad + 8K_{2}E \left[ \int_{0}^{t_{1}} \left( \left| \widetilde{y}(s)-z_{0}(s) \right|^2 + \sum\limits_{i = 1}^{\ell} \left| \widetilde{y}(\delta_i(t))-z_i(s) \right|^2 \right) ds \right] \\ = : & 8K_2(I_8+I_9). \end{align}$

(5.20)

For $I_8$ , we have

$\begin{align} I_8 \leq & (\ell+1) \int_{0}^{t_1} E\left[ \sup\limits_{0 \leq r\leq s} \left| \widetilde{X}(r)-\widetilde{y}(s) \right|^2 \right]ds. \end{align}$

(5.21)

Using (5.14) and (5.19), we have

$\begin{align} I_9 \leq & T \left\{ E \left[ \sup\limits_{0 \leq s\leq T} \left| \widetilde{y}(s)-z_{0}(s) \right|^2 \right] + \sum\limits_{i = 1}^{\ell} E \left[\sup\limits_{0 \leq s\leq T} \left| \widetilde{y}(\delta_{i}(t))-z_{i}(s) \right|^2 \right] \right\} \\ \leq & T \left( \widetilde{C}_1 \varDelta + \ell \widetilde{C}_2 \varDelta^{\gamma} \right). \end{align}$

(5.22)

(5.21), (5.22) and Gronwall's lemma imply that

$\begin{align} & E \left[ \sup\limits_{0\leq t \leq T} |\widetilde{X}(t)-\widetilde{y}(t)|^2 \right] \leq 8K_{2}T \left( \widetilde{C}_1 + \ell \widetilde{C}_2 \right) e^{8K_2T(\ell+1)} \varDelta^{\gamma}. \end{align}$

(5.23)

This inequality and Chebyshev's inequality imply (3.3).

6. Conclusions

In this article, we consider the problem of Euler–Maruyama approximate solutions of SDEs with multiple delay functions. The main result of this article is Theorem 2.1. In the theorem, we obtain the rate of convergence of approximate solutions to the exact solutions. We have applied the results to obtain confidence interval estimations for the approximate solutions. This information improves the understanding of gaps between exact solutions of SDEs with multiple delays by applying stochastic analysis and numerical solutions through simulation studies.

Acknowledgments

The authors thank Professor Yozo Tamura for his useful comments. They also thank Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript. This research was supported by JSPS KAKENHI Grant number 18K03324.

Conflict of interest

All authors declare no conflicts of interest in this paper.

References

[1]	N. Sykes, Constipation and diarrhea, In: Management of advanced disease, 4 Eds., CRC Press, 2004. http://doi.org/10.1093/med/9780199656097.003.0203
[2]	T. M. Sørensen, E. M. Vestergaard, S. K. Jensen, C. Lauridsen, S. Højsgaard, Performance and Diarrhea in piglets following weaning at seven weeks of age: challenge with E. coli O 149 and effect of dietary factors. Livest. Sci., 123 (2009), 314–321. http://doi.org/10.1016/j.livsci.2008.12.001 doi: 10.1016/j.livsci.2008.12.001
[3]	A. A. Sancak, H. C. Rutgers, C. A. Hart, R. M. Batt, Prevalence of enteropathic Escherichia coli in dogs with acute and chronic diarrhea, Vet. Rec., 154 (2004), 101–106. http://doi.org/10.1136/vr.154.4.101 doi: 10.1136/vr.154.4.101
[4]	H. Szajewska, A. Skorka, M. Dylag, Meta-analysis: Saccharomyces boulardii for treating acute Diarrhea in children, Alimentary Pharmacology and Therapeutics, 25 (2007), 257–264. http://doi.org/10.1111/j.1365-2036.2006.03202.x doi: 10.1111/j.1365-2036.2006.03202.x
[5]	A. Guarino, A. L. Vecchio, R. B. Canani, Chronic diarrhea in children, Best Pract. Res. Cl. Ga., 26 (2012), 649–661. http://doi.org/10.1016/j.bpg.2012.11.004 doi: 10.1016/j.bpg.2012.11.004
[6]	U. Navaneethan, R. A. Giannella, Definition, epidemiology, pathophysiology, clinical classification, and differential diagnosis of diarrhea, In: Diarrhea, Totowa, NJ: Humana Press, 2010, 1–31. http://doi.org/10.1007/978-1-60761-183-7_1
[7]	A. P. S. Hungin, L. Paxman, K. Koenig, J. Dalrymple, N. Wicks, J. Walmsley, Prevalence, symptom patterns and management of episodic Diarrhea in the community: a population-based survey in 11 countries, Alimentary Pharmacology and Therapeutics, 43 (2016), 586–595. http://doi.org/10.1111/apt.13513 doi: 10.1111/apt.13513
[8]	G. G. Gunsa, K. M. Rodamo, D. D. Dangiso, Determinants of acute diarrhea among children aged 6-59 months in Chiffre District, Southern Ethiopia: Unmatched case-control study, Journal of Gynecology and Obstetrics, 6 (2018), 15–25. http://doi.org/10.11648/j.jgo.20180602.11 doi: 10.11648/j.jgo.20180602.11
[9]	S. Kauchali, N. Rollins, J. Van den Broeck, Local beliefs about childhood diarrhea: importance for healthcare and research, J. Trop. Pediatrics, 50 (2004), 82–89. http://doi.org/10.1093/tropej/50.2.82 doi: 10.1093/tropej/50.2.82
[10]	E. Bonyah, G. Twagirumukiza, P. P. Gambrah, Mathematical analysis of diarrhea model with saturated incidence rate, Open J. Math. Sci., 3 (2019), 29–39. http://doi.org/10.30538/oms2019.0046 doi: 10.30538/oms2019.0046
[11]	L. Chola, J. Michalow, A. Tugendhaft, K. Hofman, Reducing diarrhea deaths in South Africa: costs and effects of scaling up essential interventions to prevent and treat Diarrhea in under-five children, BMC Public Health, 15 (2015), 394. http://doi.org/10.1186/s12889-015-1689-2 doi: 10.1186/s12889-015-1689-2
[12]	M. V. Jose, J. R. Bobadilla, Epidemiological model of diarrheal diseases and its application in prevention and control, Vaccine, 12 (1994), 109–116. http://doi.org/10.1016/0264-410x(94)90047-7 doi: 10.1016/0264-410x(94)90047-7
[13]	B. F. Iyun, E. A. Oke, Ecological and cultural barriers to treatment of childhood diarrhea in riverine areas of Ondo State, Nigeria, Social Science and Medicine, 50 (2000), 953–964. http://doi.org/10.1016/s0277-9536(99)00347-0 doi: 10.1016/s0277-9536(99)00347-0
[14]	A. Shaikh, K. S. Nisar, V. Jadhav, S. K. Elagan, M. Zakarya, Dynamical behaviour of HIV/AIDS model using fractional derivative with Mittag-Leffler kernel, Alex. Eng. J., 61 (2022), 2601–2610, http://doi.org/10.1016/j.aej.2021.08.030 doi: 10.1016/j.aej.2021.08.030
[15]	M. Farman, A. Akgül, K. S. Nisar, D. Ahmad, A. Ahmad, S. Kamangar, et al., Epidemiological analysis of fractional order COVID-19 model with Mittag-Leffler kernel, AIMS Mathematics, 7 (2022), 756–783. http://doi.org/10.3934/math.2022046 doi: 10.3934/math.2022046
[16]	K. Logeswari, C. Ravichandran, K. S. Nisar, Mathematical model for spreading of COVID-19 virus with the Mittag–Leffler kernel, Numer. Methods Partial Differential Equations, in press. http://doi.org/10.1002/num.22652
[17]	C. Ravichandran, K. Logeswari, S. K. Panda, K. S. Nisar, On new approach of fractional derivative by Mittag-Leffler kernel to neutral integro-differential systems with impulsive conditions, Chaos Soliton. Fract., 139 (2020), 110012. http://doi.org/10.1016/j.chaos.2020.110012 doi: 10.1016/j.chaos.2020.110012
[18]	S. Kumar, R. Kumar, J. Singh, K. S. Nisar, D. Kumar, An efficient numerical scheme for fractional model of HIV-1 infection of CD4+ T-cells with the effect of antiviral drug therapy, Alex. Eng. J., 59 (2020), 2053–2064. http://doi.org/10.1016/j.aej.2019.12.046 doi: 10.1016/j.aej.2019.12.046
[19]	I. Zada, M. N. Jan, N. Ali, D. Alrowail, K. S. Nisar, G. Zaman, Mathematical analysis of hepatitis B epidemic model with optimal control. Adv. Differ. Equ., 2021 (2021), 451. http://doi.org/10.1186/s13662-021-03607-2 doi: 10.1186/s13662-021-03607-2
[20]	I. Podlubny, Fractional differential equations of mathematics in science and engineering, New York, NY, USA: Academic Press, 1999.
[21]	A. Atangana, B. Dumitru, New fractional derivatives with non-local and non-singular kernel: Theory and application to heat transfer model, Therm. Sci., 20 (2016), 763–769. http://doi.org/10.2298/TSCI160111018A doi: 10.2298/TSCI160111018A
[22]	P. Ezanno, C. Fourichon, A. F. Viet, H. Seegers, Sensitivity analysis to identify key-parameters in modelling the spread of bovine viral diarrhea virus in a dairy herd, Prev. Vet. Med., 80 (2007), 49–64. http://doi.org/10.1016/j.prevetmed.2007.01.005 doi: 10.1016/j.prevetmed.2007.01.005
[23]	M. Toufik, A. Atangana, New numerical approximation of fractional derivative with non-local and non-singular kernel: application to chaotic models, Eur. Phys. J. Plus, 132 (2017), 444. http://doi.org/10.1140/epjp/i2017-11717-0 doi: 10.1140/epjp/i2017-11717-0
[24]	T. A. Biala, A. Q. M. Khaliq, A fractional-order compartmental model for the spread of the COVID-19 pandemic, Commun. Nonlinear. Sci. Numer. Simulat., 98 (2021), 105764. http://doi.org/10.1016/j.cnsns.2021.105764 doi: 10.1016/j.cnsns.2021.105764
[25]	Y. Li, Y. Chen, I. Podlubny, Mittag–Leffler stability of fractional order nonlinear dynamic systems, Automatica, 45 (2009), 1965–1969. http://doi.org/10.1016/j.automatica.2009.04.003 doi: 10.1016/j.automatica.2009.04.003

This article has been cited by:

Issam Aboutaib, Janusz Brzdęk, Lahbib Oubbi, Barrelled Weakly Köthe–Orlicz Summable Sequence Spaces, 2023, 12, 2227-7390, 88, 10.3390/math12010088

Reader Comments

Your name:*

Email:*
© 2022 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)