A Tail-Respecting Splitting Numerical Scheme for L'evy-Driven SDEs With Superlinear Drifts
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.
Ada: We're looking at the paper: "A Tail-Respecting Splitting Numerical Scheme for L'evy-Driven SDEs With Superlinear Drifts".
Emmy: Here's the gist of the paper. A Tail-Respecting Splitting Numerical Scheme for L´evy-Driven SDEs With Superlinear Drifts presents an explicit splitting numerical approximation scheme for solving multivariate stochastic differential equations driven by a multiplicative heavy-tailed Lévy process with a finite p-th moment, showing strong convergence rates that respect the finite moments of the solution.
Ada: Setting and Convergence Results The paper investigates solutions to an Itô SDE (1.1) driven by a standard Brownian motion and an independent Lévy process Z with a characteristic function defined in (1.2), establishing strong convergence rates for the numerical scheme tXn u.
Emmy: Specifically, it shows that the strong L pX-convergence holds as O(nq) for any pX P p0, p > κ-1q, which is the range where the pX-moment of the solution X is known to be finite.
Ada: Furthermore, strong uniform convergence E suptPr0,T s (Xn t - Xt)pX O(nq) is established for any pX P p0, pq.
Paper summary: Ada: So we're looking at this paper about a new numerical scheme for stochastic differential equations driven by Lévy processes that have superlinearly growing drifts. It claims they've developed an explicit splitting approximation, called the scheme tX n u, for solving these types of problems one <ref:2504.07255#pg1,the scheme tX n u>.
Emmy: It’s focused on showing strong convergence rates that respect the finite moments of the solution X <ref:2504.07255#pg1>. What this means is that if you know a certain moment of the true solution, say the p-th moment, then this numerical method gives you a predictable error rate involving n and q <ref:2504.07255#pg1>.
Kolya: The paper specifically looks at SDEs driven by a standard Brownian motion and an independent Lévy process Z <ref:2504.07255#pg1>. They establish strong L pX convergence as O of n q for any p less than κ minus one q, which is the range where that p-th moment of the solution X is known to be finite <ref:2504.07255#pg1>.
Ada: And they also establish strong uniform convergence, which is E suptPr0,T s X n t - Xt pX “ O of n q for any p less than or equal to p zero q <ref:2504.07255#pg1>. That’s a lot of information about how accurate the simulation gets over time and across different paths <ref:2504.07255#pg1>.
Emmy: And they also point out something interesting in the special case where the SDE is driven only by Brownian motion, where their numerical scheme preserves super-exponential moments of the solution <ref:2504.07255#pg1>. That's a structural property they found that holds up under certain conditions <ref:2504.07255#pg1>.
Kolya: But then they run into some trouble when the noise is driven by a Lévy process because neither Z nor X generally has all absolute moments, which complicates things <ref:2504.07255#pg1>. They propose a scheme called the explicit Lie–Trotter–Euler scheme, or tX n u, by decoupling the dynamics governed by the superlinearly growing vector field A <ref:2504.07255#pg3>.
Ada: What I find interesting is their main result, Theorem one point two, which shows that a kappa dissipative drift transforms the p-moment of the noise into approximately a pp greater than kappa minus one q moment of the solution X <ref:2504.07255#pg2>.
Emmy: And they show that this scheme tX n u preserves this tail-improving effect, which they say is delicate and not shared by other splitting schemes <ref:2504.07255#pg1>. The reverse splitting scheme, for example, doesn't preserve finite moments beyond what the integrability of Z allows <ref:2504.07255#pg1>.
Paper summary: Kolya: Their convergence rates are detailed too, including Lq convergence rates in Theorem one point four with a rate of O of n to the p minus q times chi to q over two <ref:2504.07255#pg3>. And the second main result, Theorem one point six, provides a bound on the convergence rate that incorporates this tail-improving effect <ref:2504.07255#pg3>.
Ada: They also show something about Gaussian super-exponential tails with index two transforming into solution’s super-exponential tails with index one in the continuous case, which is shown in Theorems one point seven and one point eight <ref:2504.07255#pg3>.
Emmy: That transformation is achieved using Lyapunov functions, where they show that the Lévy-driven SDE satisfies certain bounds on its moments through Itô's formula combined with localization and the Fatou Lemma to get estimates for V pXnq <ref:2504.07255#pg3>.
Kolya: The final estimate they provide, which is (four point six one), gives the required convergence rate for the moments of delta n t, which is C N to the one over h to the p minus q times chi to q n p one squared x to p q p two N to one over h q two cubed times n <ref:2504.07255#pg3>.
Ada: It sounds like they've done a lot of work on making sure the numerical approximation actually respects those finite moment conditions they are looking for <ref:2504.07255#pg1>. The explicit Euler scheme they started with has issues, like exploding in finite time and giving out non-real numbers with positive probability <ref:2504.07255#pg3>.
Emmy: They fixed that by proposing the tamed Euler scheme, which is designed so that the drift term stays bounded to prevent those non-real number problems <ref:2504.07255#pg3>. But they noted that for Lévy-driven components, neither Z nor X always have all absolute moments <ref:2504.07255#pg1>.
Kolya: So the explicit scheme tX n u they propose is built by decoupling the deterministic dynamics governed by the superlinearly growing vector field A from other components one <ref:2504.07255#pg1>. This scheme deviates a little from standard Lie–Trotter splitting because it uses a one-step Euler approximation of an SDE one <ref:2504.07255#pg1>.
Ada: And they realized that this specific construction, tX n u, is what preserves that tail-improving effect mentioned earlier, which is what makes it different from the tamed Euler scheme or the other alternative splitting scheme <ref:2504.07255#pg1>.
Paper summary: Emmy: The real power here seems to come from combining estimates derived from Theorems one point two and one point three, which show that a kappa dissipative drift transforms the p-moment of the noise into approximately a pp greater than kappa minus one q moment of the solution X <ref:2504.07255#pg3>.
Kolya: The convergence rates are derived from Lemma four point three and Lemma four point five, leading to that final estimate which shows the error is bounded by C N to the one over h to the p minus q times chi to q n p one squared x to p q p two N to one over h q two cubed times n <ref:2504.07255#pg3>.
Ada: In the continuous case, Theorems one point seven and one point eight show that a kappa dissipative drift transforms Gaussian super-exponential tails with index two into solution’s super-exponential tails with index one <ref:2504.07255#pg3>. That’s a specific kind of tail behavior they are controlling <ref:2504.07255#pg3>.
Emmy: They prove those theorems using Lyapunov functions, which show that the Lévy-driven SDE satisfies certain bounds on its moments through Itô's formula combined with localization and the Fatou Lemma to obtain estimates for V pXnq <ref:2504.07255#pg3>.
Kolya: That leads to the estimate one over one forty, which is obtained analogously with the help of those bounds <ref:2504.07255#pg3>.
Ada: So, in short, this paper presents an explicit splitting numerical approximation scheme for solving multivariate SDEs driven by a multiplicative heavy-tailed Lévy process that has a finite p-th moment <ref:2504.07255#pg1>. It shows strong convergence rates that respect the finite moments of the solution X <ref:2504.07255#pg1>.
Emmy: The paper, "A Tail-Respecting Splitting Numerical Scheme for L'evy-Driven SDEs With Superlinear Drifts," by Olga Aryasova, Oleksii Kulyk, and Ilya Pavlyukevich, suggests a way to numerically approximate these complex equations using a scheme that respects the heavy tails of the noise <ref:2504.07255#pg1>.
Kolya: For someone who only listens to this show about SDEs, it means that when you are modeling systems with heavy-tailed noise, you need a numerical method that doesn't just approximate the average behavior but one that actually respects the finite moments of the solution <ref:2504.07255#pg1>.
Ada: It means we can get stronger error bounds than what might be achievable with simpler methods when dealing with these superlinear drifts and heavy-tailed noise <ref:2504.07255#pg1>.
Emmy: The paper, "A Tail-Respecting Splitting Numerical Scheme for L'evy-Driven SDEs With Superlinear Drifts," by Olga Aryasova, Oleksii Kulyk, and Ilya Pavlyukevich, suggests a way to numerically approximate these complex equations using a scheme that respects the heavy tails of the noise <ref:2504.07255#pg1>.
Conclusion: Ada: So we're wrapping up this session on this paper, "A Tail-Respecting Splitting Numerical Scheme for L'evy-Driven SDEs With Superlinear Drifts."
Emmy: It’s about a new way to numerically solve these complicated stochastic differential equations where the noise has heavy tails and the drift term grows really fast.
Kolya: The main point here is that they built a specific splitting method, this scheme tX n u, that actually manages to keep track of the finite moments of the solution even when there are superlinear drifts involved.
Ada: Exactly. They found this specific structure in how they split the dynamics that lets them control those tail behaviors better than other methods they tested.
Emmy: The authors are Olga Aryasova, Oleksii Kulyk, and Ilya Pavlyukevich. They're working on making sure our numerical tools can handle these heavy-tailed noise processes more accurately than before.
Kolya: What this means for the world is that if you're modeling something like a financial market or some physical system with extreme jumps in the noise, this scheme gives you a way to get reliable results based on the actual expected moments of what's happening.
Ada: It moves us away from just getting an average behavior and toward understanding how big deviations from that average actually behave.
Emmy: They’re showing that even with these tricky superlinear drifts, you can still get strong convergence rates tied directly to the moment conditions of the solution itself.
Kolya: So, it’s about matching the numerical accuracy to the statistical properties of what you're trying to model, which is a pretty solid approach for complex systems.
Ada: It really shows how important it is to get those details right in numerical simulations when dealing with processes that have such extreme jumps.
Emmy: And they’ve shown this works not just in theory but also that it respects some very specific tail transformations, like turning index two tails into index one tails in the continuous case.
Kolya: That level of control over the tails is what makes this work interesting; it’s not just about getting a general approximation anymore.
Ada: It sets a new benchmark for how we can design algorithms to solve these kinds of difficult stochastic problems where the noise isn't well-behaved in the traditional sense.
Institute of Mathematics, Friedrich Schiller University Jena · Institute of Geophysics, National Academy of Sciences of Ukraine · Igor Sikorsky Kyiv Polytechnic Institute
math.PR, cs.NA, math.NA
Submitted: 2025-04-09
Updated: 2026-11-19
Comments: 43 pages
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 83/100
The gist: A Tail-Respecting Splitting Numerical Scheme for L´evy-Driven SDEs With Superlinear Drifts presents an explicit splitting numerical approximation scheme for solving multivariate stochastic…
Key concepts
- Lévy-Driven SDEs
- These are stochastic differential equations where the randomness comes from a Lévy process, which can have jumps (like sudden, large changes) in addition to continuous movement. The paper focuses on cases where these processes also have superlinear drifts, meaning the deterministic part of the equation grows faster than linearly.
- Tail-Respecting Splitting Scheme
- This is a numerical method that solves complex SDEs by breaking them into simpler parts (splitting) and combining their solutions. The 'tail-respecting' aspect means the scheme is specifically constructed so that its accuracy and convergence rates are tied to the finite moments of the solution, ensuring it handles heavy tails correctly.
- Tail-Improving Effect
- This key finding shows that a specific type of drift term acts like a filter. It takes noise with very heavy tails (which have infinite moments) and transforms them into solution tails that have finite moments up to a certain order, improving the numerical stability and convergence properties.
- Superlinear Drifts
- These refer to deterministic terms in the SDE whose growth rate is faster than linear. In numerical methods, these terms are challenging because they can cause solutions to explode or become unstable if not handled carefully. The scheme proposed here uses a specific decoupling technique to manage this complexity.
Terminology
Summary
A Tail-Respecting Splitting Numerical Scheme for L´evy-Driven SDEs With Superlinear Drifts presents an explicit splitting numerical approximation scheme for solving multivariate stochastic differential equations driven by a multiplicative heavy-tailed Lévy process with a finite p-th moment, showing strong convergence rates that respect the finite moments of the solution. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
Setting and Convergence Results
The paper investigates solutions to an Itô SDE (1.1) driven by a standard Brownian motion and an independent Lévy process Z with a characteristic function defined in (1. p. 0 1 Setting and results
In this paper we focus on strong numerical approximations of solutions of an Itˆo SDE with multiplicative
noise
Xt “ x `
ż t
0
ApXsq ds `
ż t
0
apXsq ds `
ż t
0
bpXsq dBs `
ż t
0
cpXs´q dZs, x P R
d
, t P r0, 8q, (1. 1)
driven by a d-dimensional standard Brownian motion B, d P N, and an independent d-dimensional L´evy
process Z with the characteristic function
Ee
ixλ,Zty “ exp ´
t
ż
Rd
`
e
ixλ,zy ´ 1 ´ ixλ, zyIp}z} ď 1q
˘
νpdzq
¯
, λ P R
d
, t P r0, 8q. In both cases we determine the convergence rates γ and δ. In the special case of SDEs driven solely by a Brownian motion, our numerical scheme preserves
super-exponential moments of the solution. A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite. 4 will be given in Section 4. Within the framework of this Theorem, the uniform
convergence of E suptPr0,Ts
}Xn
t ´Xt}
q
can also be established, although with a slightly weaker approximation
rate. In both cases we determine the convergence rates γ and δ. In the special case of SDEs driven solely by a Brownian motion, our numerical scheme preserves
super-exponential moments of the solution.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Scheme Construction and Limitations
The numerical scheme tXn u is realized as a combination of an Euler method with a Lie–Trotter type splitting technique. The explicit Euler scheme (1. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate. However in practice, computer implementations of the explicit Euler scheme (1. 12) XE,n may explode in
finite time and yield NaN (“Not a Number”) outputs with positive probability. Addressing this question is far from straightforward, as intuitive candidates — such as the explicit Euler
or the tamed/truncated explicit Euler schemes — fail to accurately reproduce the moment behaviour in the
range pX P pp, p ` κ ´ 1q. In this paper, we propose a novel explicit approximation scheme, which is tail-respecting in the sense
described above. As explained in Section 1. 1, the drift term in the tamed Euler scheme is bounded, the therefore its
sample paths remain finite. First, the presence of (heavy) jumps
causes NaN outputs from the explicit Euler scheme to arise more frequently than in the Gaussian case; see
Example 1. Second, unlike in the Gaussian setting, neither the driving L´evy process Z, nor the true
solution X generally possess all absolute moments. A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite. More precisely, the solution X is approximated in two steps by decoupling the deterministic dynamics
governed by the superlinearly growing vector field A from the remaining components. The resulting scheme
is as follows.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Tail-Improving Effect and Scheme Comparison
The main result, Theorem 1.2, demonstrates that a κ-dissipative drift transforms the p-moment of the noise into (approximately) a pp > κ-1q-moment of the solution X. p. 3 1) satisfies
sup
tPr0,8q
E}Xt}
pX ďC 1 ` }x}
pX , x P R
d
. 17)
This theorem reveals a general phenomenon: a κ-dissipative drift transforms the p-moment of the noise
into (approximately) a pp ` κ ´ 1q-moment of the solution X. These methods require increased computational effort and will not be considered further in this
work. An alternative approach, proposed in [21], is the tamed explicit Euler scheme:
X
TE,n
0 “ x,
X
TE,n
t
n
k`1
“ X
TE,n
t
n
k
`
ArpX
TE,n
t
n
k
qhn
1 ` }ArpX
TE,n
t
n
k
q}hn
` bpX
TE,n
t
n
k
qpBt
n
k`1
´ Bt
n
k
q, k “ 0, . 23) preserves the same tail-improving effect as the original
SDE (1. It is important to emphasize that this property is quite delicate and is not shared by the tamed
Euler scheme or an alternative splitting scheme, which we briefly discuss below. 6,
the splitting scheme Xn preserves the uniform convergence of individual moments, albeit with a slower
convergence rate for moments pX approaching the critical value p ` κ ´ 1 “ 3; see Table 6. In contrast,
10 26)
Therefore, (1. 25) does not preserve the finite moments of X beyond the order allowed by the integrability
of Z.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Convergence Rates and Final Estimates
The paper establishes several convergence rates for the scheme (1.19), including Lq-convergence rates in Theorem 1. p. 5 The following theorem, which is the second main result of the paper, is specific to the scheme (1. 23) and combines the tail-improving effect with convergence rates. The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate. 5 focus solely on the moments of the L´evy process Z, we strongly believe that
analogous results should hold for the tamed Euler scheme as well as the reverse splitting scheme (1. The following theorem, which is the second main result of the paper, is specific to the scheme (1. The first step focuses on
the L
q
-convergence rates for q P p0, pq. This step does not require κ-dissipativity and does not involve the
tail-improving effect. ,
limnÑ8 E}X
E,n
T
}
r “ 8. In the continuous case, c ” 0, the deficiencies of the explicit Euler method can be addressed in two main
ways. 37)
Theorem 1. 7 demonstrates that a κ-dissipative drift transforms Gaussian super-exponential tails with
index 2 into solution’s super-exponential tails with index 1 ` κ ą 2. 38)
and for any T P r0, 8q and λ P p0,
Λ
2
q
lim sup
nÑ8
E
”
sup
tPr0,Ts
e
λ
1`κ
}Xn
t
}
κ`1
ı
ďC e
λ
1`κ
}x}
1`κ
, x P R
d
. 39)
The final theorem in this section provides the rate of convergence of our splitting scheme, accounting for
the super-exponential tails with index 1 ` κ ą 2.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Continuous Case Analysis
In the continuous case, Theorems 1.7 and 1.8 are proven using Lyapunov functions, showing that the Lévy-driven SDE (1. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite. The difference between the splitting and the reverse splitting
schemes then is determined by the choice of steps after which the values of the scheme are taken out:
deterministic for the splitting scheme, and stochastic for the reverse splitting one. Since each deterministic
step keeps corresponding values of the two-step scheme under the threshold Kn, the terminal value of the
reverse splitting scheme satisfies E|Xpn
T
|
pX “ `8 for pX P r1, 8q only because the corresponding moments
of the increment of Z on the very last stochastic step are infinite. The difference between the splitting and the reverse splitting
schemes then is determined by the choice of steps after which the values of the scheme are taken out:
deterministic for the splitting scheme, and stochastic for the reverse splitting one. Since each deterministic
step keeps corresponding values of the two-step scheme under the threshold Kn, the terminal value of the
reverse splitting scheme satisfies E|Xpn
T
|
pX “ `8 for pX P r1, 8q only because the corresponding moments
of the increment of Z on the very last stochastic step are infinite. 38)
and for any T P r0, 8q and λ P p0,
Λ
2
q
lim sup
nÑ8
E
”
sup
tPr0,Ts
e
λ
1`κ
}Xn
t
}
κ`1
ı
ďC e
λ
1`κ
}x}
1`κ
, x P R
d
. 39)
The final theorem in this section provides the rate of convergence of our splitting scheme, accounting for
the super-exponential tails with index 1 ` κ ą 2. 38)
and for any T P r0, 8q and λ P p0,
Λ
2
q
lim sup
nÑ8
E
”
sup
tPr0,Ts
e
λ
1`κ
}Xn
t
}
κ`1
ı
ďC e
λ
1`κ
}x}
1`κ
, x P R
d
. 39)
The final theorem in this section provides the rate of convergence of our splitting scheme, accounting for
the super-exponential tails with index 1 ` κ ą 2. In columns 3–7 of Table 1, we present the empirical absolute moments of the
tamed Euler scheme X
TE,n
T which are calculated as follows. Let tX
TE,n
T
piqu1ďiďN be the empirical results
obtained in N independent series of simulations. In columns 3–7 of Table 1, we present the empirical absolute moments of the
tamed Euler scheme X
TE,n
T which are calculated as follows. Let tX
TE,n
T
piqu1ďiďN be the empirical results
obtained in N independent series of simulations.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
References
[1] A. Chechkin, V. Gonchar, J. Klafter, R. Metzler, L. Tanatarov, Stationary states of non-linear oscillators driven by L´evy noise. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10].Where the paper says it
[2] A. V. Chechkin, V. Y. Gonchar, J. Klafter, R. Metzler, L´evy flights in a steep potential well. p. 5 31)
Since Theorem 1. 5 focus solely on the moments of the L´evy process Z, we strongly believe that
analogous results should hold for the tamed Euler scheme as well as the reverse splitting scheme (1.Where the paper says it
[3] A. V. Chechkin, V. Y. Gonchar, J. Klafter, R. Metzler, Natural cutoff in L´evy flights caused by dissipative nonlinearity. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[4] B Dybiec, E Gudowska-Nowak, P Hänggi, Escape driven by α-stable white noises. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10].Where the paper says it
[5] B Dybiec, I M Sokolov, A V Chechkin, Stationary states in single-well potentials under symmetric L´evy noises. p. 3 tamed Euler scheme is bounded from above by 1, the conditional expectation of the scheme’s q-th moment
diverges:
E
”
}X
TE,n
t
n
k`1
}
q
ˇ
ˇ
ˇ
Ft
n
k
ı
ěC ´}X
TE,n
t
n
k
}
q ´ 1 ´ }b}
qE}Bt
n
k`1
´ Bt
n
k
}
q ` c
q
˚E}Zt
n
k`1
´ Zt
n
k
}
q
ěC ´}X
TE,n
t
n
k
}
q ´ 1 ` E}Zt
n
k`1
´ Zt
n
k
}
q “ `8 a. 16)
This explains why the works [28, 29, 30] consider p-integrable L´evy noises with p P r2, 8q and establish
convergence in the L
r
-sense for r P p0, pq only, see, e.Where the paper says it
[6] B Dybiec, M Zawi´slak, Escape from the potential well: accelerating by shaping and noise tuning.
[7] B Lindner, The diffusion coefficient of nonlinear Brownian motion. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[8] B Lindner, Diffusion coefficient of a Brownian particle with a friction function given by a power law. p. 6 Splitting methods for SDEs with cubic nonlinearity and additive noise were studied in [42]; see
also the discussion in Example 1. More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications.Where the paper says it
[9] B Lindner, Diffusion of particles subject to nonlinear friction and a colored noise. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[10] A Kulik, I Pavlyukevich, Non-Gaussian limit theorem for non-linear Langevin equations driven by L´evy noise. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10].Where the paper says it
[11] S Borovkova, F J Permana, Modelling electricity prices by the potential jump-diffusion. p. 1 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10]. In financial mathematics, such SDEs are also used for the simulation of energy prices, see [11, 12].Where the paper says it
[12] S Borovkova, F J Permana, I Pavlyukevich, Modeling electricity prices by potential L´evy diffusions. p. 1 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10]. In financial mathematics, such SDEs are also used for the simulation of energy prices, see [11, 12].Where the paper says it
[13] I Gy¨ongy, N Krylov, Existence of strong solutions for Itˆo’s stochastic equations via approximations. p. 0 2020 Mathematics Subject Classification: 60H35˚, 60G51, 60H10, 65C05, 65C30
1 Main results
1. 1 Setting and results
In this paper we focus on strong numerical approximations of solutions of an Itˆo SDE with multiplicative
noise
Xt “ x `
ż t
0
ApXsq ds `
ż t
0
apXsq ds `
ż t
0
bpXsq dBs `
ż t
0
cpXs´q dZs, x P R
d
, t P r0, 8q, (1.Where the paper says it
[14] I Gy¨ongy, A note on Euler’s approximations. p. 2 , in the absence of jumps, the SDE (1. 1) with the superlinearly increasing drift and its Euler
approximations have been extensively studied in the literature, see, e.Where the paper says it
[15] N V Krylov, On Kolmogorov’s equations for finite dimensional diffusions. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[16] X Mao, Stochastic Differential Equations and Applications, 2nd Edition, Woodhead Publishing. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[17] M Hutzenthaler, A Jentzen, P E Kloeden, Strong and weak divergence in finite time of Euler’s method for stochastic differential equations with non-globally Lipschitz continuous coefficients. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[18] P E Kloeden, E Platen, Numerical Solution of Stochastic Differential Equations. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[19] H Schurz, Numerical analysis of SDEs without tears. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[20] D J Higham, X Mao, A M Stuart, Strong convergence of Euler-type methods for nonlinear stochastic differential equations. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[21] M Hutzenthaler, A Jentzen, P E Kloeden, Strong convergence of an explicit numerical method for SDEs with nonglobally Lipschitz continuous coefficients. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[22] M Hutzenthaler, A Jentzen, Numerical approximations of stochastic differential equations with nonglobally Lipschitz continuous coefficients, Vol. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[23] S Sabanis, A note on tamed Euler approximations, Electronic Communications in Probability 18 (47). p. 4 It is important to emphasize that this property is quite delicate and is not shared by the tamed
Euler scheme or an alternative splitting scheme, which we briefly discuss below. It is well known, that the splitting order the in Lie–Trotter approximations can be arbitrary.Where the paper says it
[24] S Sabanis, Euler approximations with varying coefficients: the case of superlinearly growing diffusion coefficients. p. 3 For background on operator splitting and related product formula
approximations, we refer the reader to the classical works [35, 36, 37] and the recent review [38]. More precisely, the solution X is approximated in two steps by decoupling the deterministic dynamics
governed by the superlinearly growing vector field A from the remaining components.Where the paper says it
[25] X Mao, The truncated Euler–Maruyama method for stochastic differential equations. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[26] X Mao, Convergence rates of the truncated Euler–Maruyama method for stochastic differential equations. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[27] X Li, X Mao, G Yin, Explicit numerical approximations for stochastic differential equations in finite and infinite horizons: truncation methods. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[28] K Dareiotis, C Kumar, S Sabanis, On tamed Euler approximations of SDEs driven by L´evy noise with applications to delay equations. p. 0 1 Setting and results
In this paper we focus on strong numerical approximations of solutions of an Itˆo SDE with multiplicative
noise
Xt “ x `
ż t
0
ApXsq ds `
ż t
0
apXsq ds `
ż t
0
bpXsq dBs `
ż t
0
cpXs´q dZs, x P R
d
, t P r0, 8q, (1. 1)
driven by a d-dimensional standard Brownian motion B, d P N, and an independent d-dimensional L´evy
process Z with the characteristic function
Ee
ixλ,Zty “ exp ´
t
ż
Rd
`
e
ixλ,zy ´ 1 ´ ixλ, zyIp}z} ď 1q
˘
νpdzq
¯
, λ P R
d
, t P r0, 8q.Where the paper says it
[29] C Kumar, S Sabanis, On tamed Milstein schemes of SDEs driven by L´evy noise. p. 6 Splitting methods for SDEs with cubic nonlinearity and additive noise were studied in [42]; see
also the discussion in Example 1. More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications.Where the paper says it
[30] C Kumar, S Sabanis, On explicit approximations for L´evy driven SDEs with super-linear diffusion coefficients. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10].Where the paper says it
[31] S Deng, W Fei, W Liu, X Mao, The truncated EM method for stochastic differential equations with Poisson jumps. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[32] T T Kieu, D T Luong, H L Ngo, N K Tran, Strong convergence in infinite time interval of tamedadaptive Euler–Maruyama scheme for L´evy-driven SDEs with irregular coefficients. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[33] C Kelly, G J Lord, F Sun, Strong convergence of a class of adaptive numerical methods for SDEs with jumps. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[34] A Kulik, I Pavlyukevich, Moment bounds for dissipative semimartingales with heavy jumps. p. 4 3 are derived from the general result in Theorem 2. 12 of [34], which provides sharp bounds on the
tails of Itˆo semimartingales with dissipative drifts and heavy-tailed jumps.Where the paper says it
[35] H F Trotter, On the product of semi-groups of operators. p. 3 The scheme combines the Lie–Trotter operator-splitting method for semigroup approximation with an Euler-type discretization. For background on operator splitting and related product formula
approximations, we refer the reader to the classical works [35, 36, 37] and the recent review [38].Where the paper says it
[36] G Strang, On the construction and comparison of difference schemes. p. 9 It is informative to consider both the splitting and the reverse splitting schemes as by-products of one
(hidden) two-step scheme, where the “deterministic” steps governed by the flow Φ are interchanged with the
“stochastic” Euler-type approximation steps. The difference between the splitting and the reverse splitting
schemes then is determined by the choice of steps after which the values of the scheme are taken out:
deterministic for the splitting scheme, and stochastic for the reverse splitting one.Where the paper says it
[37] P R Chernoff, Product Formulas, Nonlinear Semigroups, and Addition of Unbounded Operators. p. 3 The scheme combines the Lie–Trotter operator-splitting method for semigroup approximation with an Euler-type discretization. For background on operator splitting and related product formula
approximations, we refer the reader to the classical works [35, 36, 37] and the recent review [38].Where the paper says it
[38] M B Vovchanskyi, A quick probability-oriented introduction to operator splitting methods. p. 3 In this paper, we propose a novel explicit approximation scheme, which is tail-respecting in the sense
described above. The scheme combines the Lie–Trotter operator-splitting method for semigroup approximation with an Euler-type discretization. In this paper, we propose a novel explicit approximation scheme, which is tail-respecting in the sense
described above. The scheme combines the Lie–Trotter operator-splitting method for semigroup approximation with an Euler-type discretization.Where the paper says it
Where the paper says it
[39] I Pavlyukevich, O Aryasova, A Chechkin, O Kulyk, How to simulate L´evy flights in a steep potential: An explicit splitting numerical scheme. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[40] J C Butcher, Numerical methods for ordinary differential equations in the 20th century. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[41] E Hairer, C Lubich, Numerical solution of ordinary differential equations, in: N J Higham (Ed. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite. 37, Kyiv 03056, Ukraine; oaryasova@gmail. com
†Wroclaw University of Science and Technology, Faculty of Pure and Applied Mathematics, Wybrze´ze Wyspia´nskiego Str.Where the paper says it
Where the paper says it
[42] E Buckwar, A Samson, M Tamborrino, I Tubikanec, A splitting method for SDEs with locally Lipschitz drift: Illustration on the FitzHugh-Nagumo model. p. 0 The scheme tX
n
u is realized as a combination of a well-known Euler method with a Lie–Trotter type
splitting technique. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate.Where the paper says it
[43] P Pilipovic, A Samson, S Ditlevsen, Parameter estimation in nonlinear multivariate stochastic differential equations based on splitting schemes. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[44] E Hausenblas, T A Randrianasolo, M Thalhammer, Theoretical study and numerical simulation of pattern formation in the deterministic and stochastic Gray–Scott equations. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[45] A Kulik, Ergodic Behavior of Markov Processes. With Applications to Limit Theorems. p. 7 fpxq “ 1
πpx2`1q
. In order to estimate the accuracy of our method we compare the simulated empirical moments with the
theoretically known moments for the limit (stationary) distribution of X.Where the paper says it
[46] A Dubkov, B Spagnolo, Langevin approach to L´evy flights in fixed potentials: Exact results for stationary probability distributions.
[47] A R Humphries, A M Stuart, Runge–Kutta methods for dissipative and gradient dynamical systems. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10].Where the paper says it
[48] F Bouchut, F James, S Mancini, Uniqueness and weak stability for multi-dimensional transport equations with one-sided Lipschitz coefficient. p. 6 More recently, [43] investigated various splitting schemes for
SDEs with A satisfying the one-sided Lipschitz condition H
Lip`
A , linear drift coefficient ap¨q, and constant
diffusion coefficient bp¨q, with particular emphasis on statistical applications. In [44], splitting techniques were
employed in the numerical analysis of pattern formation for nonlinear Gray–Scott equations.Where the paper says it
[49] L Arnold, P Imkeller, Y Wu, Reduction of deterministic coupled atmosphere–ocean models to stochastic ocean models: a numerical case study of the Lorenz–Maas system.
[50] E Pardoux, A R˘a¸scanu, Stochastic Differential Equations. p. 0 A Tail-Respecting Splitting Numerical Scheme
for L´evy-Driven SDEs With Superlinear Drifts
Olga Aryasova∗
, Oleksii Kulyk† and Ilya Pavlyukevich‡
June 16, 2026
Abstract
We present an explicit splitting numerical approximation scheme, denoted by tX
n
u, for the effective
simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly
growing κ-dissipative drift, where κ ą 1, driven by a multiplicative heavy-tailed L´evy process that has a
finite p-th moment, with p ą 0. We show that the strong L
pX -convergence suptPr0,T s E}X
n
t ´ Xt}
pX “
Oph
γ
nq holds for any pX P p0, p ` κ ´ 1q, which is exactly the range where the pX-moment of the solution
X is known to be finite.Where the paper says it
[51] D D Bainov, P S Simeonov, Integral inequalities and applications, Vol. p. 1 1) is a non-local operator, it is in general difficult
to determine analytic expressions for characteristics of X like moments, first exit probabilities etc. Therefore,
the numerical approximation of X is of utmost importance in applications. 1) is a non-local operator, it is in general difficult
to determine analytic expressions for characteristics of X like moments, first exit probabilities etc. Therefore,
the numerical approximation of X is of utmost importance in applications.Where the paper says it
Where the paper says it
[52] D Applebaum, L´evy Processes and Stochastic Calculus, 2nd Edition, Vol. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10]. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate. 2020 Mathematics Subject Classification: 60H35˚, 60G51, 60H10, 65C05, 65C30
1 Main results
1.Where the paper says it
Where the paper says it
[53] K Sato, L´evy Processes and Infinitely Divisible Distributions, Vol. p. 1 SDEs of the form (1. 1) attracted considerable attention in the study of dissipative systems driven by
α-stable L´evy processes, see [1, 2, 3, 4, 5, 6] or non-linear randomly perturbed friction processes [7, 8, 9, 10]. Keywords: stochastic differential equation, heavy tails, splitting method, explicit numerical scheme, strong
convergence, superlinear drift, dissipative drift, one-sided Lipschitz condition, convergence rate. 2020 Mathematics Subject Classification: 60H35˚, 60G51, 60H10, 65C05, 65C30
1 Main results
1.Where the paper says it
Where the paper says it
[38] M B Vovchanskyi, A quick probability-oriented introduction to operator splitting methods.
How the training signal is built
The scheme (1.19) is constructed by decoupling the deterministic dynamics governed by the superlinearly growing vector field A from other components, resulting in a scheme that deviates slightly from Lie–Trotter splitting by employing a one-step Euler approximation of an SDE. p. 3 More precisely, the solution X is approximated in two steps by decoupling the deterministic dynamics
governed by the superlinearly growing vector field A from the remaining components. The resulting scheme
is as follows. We return to this issue in Example 1. As is common in such cases, it will be more convenient to work with the continuous time version of
(1. 22)
The continuous time version of (1. 19) is then given by:
Xn
0
:“ x,
Y
n
t
:“ X
pnq
t ` apX
pnq
t
qδ
n
t ` bpX
pnq
t
qB
n
t ` cpX
pnq
t
qZ
n
t
,
Xn
t
:“ Φpδ
n
t
, Y n
t
q, t P r0, Ts. In the practical implementation of our splitting scheme (1. 19) we use the explicit formula for the solution
of the ODE Φ9 “ ´Φ
3
, which leads to the result that
Φpt, xq “ x
?Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the gains come from
The gains in convergence rates stem from a combination of estimates derived from Theorems 1.2 and 1.3, which show that a κ-dissipative drift transforms the p-moment of the noise into (approximately) a pp > κ-1q-moment of the solution X. p. 3 1) satisfies
sup
tPr0,8q
E}Xt}
pX ďC 1 ` }x}
pX , x P R
d
. 17)
This theorem reveals a general phenomenon: a κ-dissipative drift transforms the p-moment of the noise
into (approximately) a pp ` κ ´ 1q-moment of the solution X. Additionally, for any pX P p0, pq we establish strong uniform convergence:
E suptPr0,T s
}X
n
t ´ Xt}
pX “ Oph
δ
nq. In both cases we determine the convergence rates γ and δ. 38)
and for any T P r0, 8q and λ P p0,
Λ
2
q
lim sup
nÑ8
E
”
sup
tPr0,Ts
e
λ
1`κ
}Xn
t
}
κ`1
ı
ďC e
λ
1`κ
}x}
1`κ
, x P R
d
. 39)
The final theorem in this section provides the rate of convergence of our splitting scheme, accounting for
the super-exponential tails with index 1 ` κ ą 2. 37)
Theorem 1. 7 demonstrates that a κ-dissipative drift transforms Gaussian super-exponential tails with
index 2 into solution’s super-exponential tails with index 1 ` κ ą 2.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Continuous Case Proof
The proof of Theorems 1.7 and 1.8 in the continuous case is based on Lyapunov function estimates, where the Lévy-driven SDE (1. p. 0 1 Setting and results
In this paper we focus on strong numerical approximations of solutions of an Itˆo SDE with multiplicative
noise
Xt “ x `
ż t
0
ApXsq ds `
ż t
0
apXsq ds `
ż t
0
bpXsq dBs `
ż t
0
cpXs´q dZs, x P R
d
, t P r0, 8q, (1. 1)
driven by a d-dimensional standard Brownian motion B, d P N, and an independent d-dimensional L´evy
process Z with the characteristic function
Ee
ixλ,Zty “ exp ´
t
ż
Rd
`
e
ixλ,zy ´ 1 ´ ixλ, zyIp}z} ď 1q
˘
νpdzq
¯
, λ P R
d
, t P r0, 8q. 16) combined with the estimate (2. 1) for the flow Φ yields
E
”
}Xpn
t
n
k`1
}
q
ˇ
ˇ
ˇ
Ft
n
k
ı
ěC ´}Φpphn, Xpn
t
n
k
q}q ´ }a}
qh
q
n ´ }b}
qE}Bt
n
k`1
´ Bt
n
k
}
q ` c
q
˚E}Zt
n
k`1
´ Zt
n
k
}
q
ěC ´}Xpn
t
n
k
}
q ´ 1 ` E}Zt
n
k`1
´ Zt
n
k
}
q “ `8 a. In the practical implementation of our splitting scheme (1. 19) we use the explicit formula for the solution
of the ODE Φ9 “ ´Φ
3
, which leads to the result that
Φpt, xq “ x
? The Euler scheme XE,n converges in probability to the true solution uniformly in t on bounded intervals,
namely, for each fixed initial value x P R
d
, any T P r0, 8q and any ε ą 0
limnÑ8
P
´
max
0ďkhďT
}X
E,n
kh ´ Xkh} ą ε
¯
“ 0, (1. 13)
see Theorem 2.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Explicit numerical scheme for superlinear drift SDEs: The scheme tXn u can be used for effective simulation of solutions X to multivariate stochastic differential equations with a superlinearly growing kappa-dissipative drift, where kappa is not equal to 1 and driven by a multiplicative heavy-tailed Levy process with a finite p-th moment.
-
Moment convergence guarantees: The scheme establishes strong LpX convergence suptPr0,T s E Xn t ´ Xt for any pX P p0, p < kappa - 1q, which is the range where the LpX-moment of the solution X is known to be finite.
-
Uniform convergence rates: For any pX P p0, pq we establish strong uniform convergence E suptPr0,T s E Xn t ´ Xt for any epsilon ą 0 with a rate of Oph delta nq.
-
Tail-respecting approximation: The explicit Lie-Trotter-Euler scheme (1.23) preserves the same tail-improving effect as the original SDE (1.1), meaning it satisfies Theorem 1.3, which is important because
this property is quite delicate and is not shared by the tamed Euler scheme or an alternative splitting scheme.
-
Convergence rate for Lq moments: For q P p0, pq there is convergence sup tPr0,Ts E Xn t ´ Xt for any epsilon ą 0 with a convergence rate of delta n p1 q.
-
Uniform moment convergence rates: There is uniform convergence E sup tPr0,Ts E Xn t ´ Xt for any epsilon ą 0 with a rate of delta n p1 q4 1/2.
-
Super-exponential tail preservation in the continuous case: For c=0, a kappa-dissipative drift transforms Gaussian super-exponential tails with index 2 into solution’s super-exponential tails with index 1, as shown by Theorem 1.7 and Theorem 1.8 for the continuous case.
-
Convergence rate for super-exponential moments: The splitting scheme (1.23) has a convergence rate of delta n p1 q, which is related to the tail-improving effect via Theorem 1.6, with an upper bound on the exponent derived from maximizing the exponent over q P p0, pq and epsilon.
-
Moment bounds for heavy-tailed jumps: The scheme provides estimates for the moments of Delta n of order q where q P p0, pq in Corollary 4.4, showing convergence rates dependent on terms involving h p'q and chi q n p1.
-
Analysis of reverse splitting scheme failure: The reverse splitting scheme (1.51) fails to preserve moments uniformly over time because
the difference between the splitting and the reverse splitting schemes becomes particularly apparent when the convergence of empirical moments is examined uniformly in time.
Abstract
We present an explicit numerical approximation scheme, denoted by X n, for the effective simulation of solutions X to a multivariate stochastic differential equation (SDE) with a superlinearly growing κ-dissipative drift, where κ>1, driven by a multiplicative heavy-tailed Lévy process that has a finite p-th moment, with p>0. We show that the strong L p X-convergence t in[0,T] E |X n t-X t| p X= O (h n γ) holds for any p X in (0,p+κ-1), which is exactly the range where the p X-moment of the solution is known to be finite. Additionally, for any p X in (0,p) we establish strong uniform convergence: E t in[0,T] |X n t-X t| p X= O (h n δ). In both cases we determine the convergence rates γ and δ. In the special case of SDEs driven solely by a Brownian motion, our numerical scheme preserves super-exponential moments of the solution. The scheme X n is realized as a combination of a well-known Euler method with a Lie-Trotter type splitting technique.
Related papers
- Rank-One HCIZ Integrals and Multiplicative Convolutions
- Repeated averaging on random d-regular graphs
- Precise cover times for branching random walks on Hamming graphs: (iterated) logarithmic corrections
- Markov Chain CLTs: Resolving Open Problems
- Marked stable L'evy processes in Brownian local time
- A Sharp Refinement of Samuels' Inequality via Cap Moves