Core Concepts
Developing explicit Milstein-type schemes for Ait-Sahalia type models with positivity preservation.
Abstract
The article introduces novel explicit Milstein-type schemes for Ait-Sahalia type models in mathematical finance.
Focuses on unconditionally positivity-preserving approximations with mean-square convergence.
Incorporates corrective mapping and implicitness to tackle difficulties in numerical approximations.
Theoretical guarantees support optimal complexity of Multilevel Monte Carlo method.
Numerical experiments validate theoretical findings.
Introduction
SDEs applications in various disciplines due to lack of analytical solutions.
Interest in numerical counterparts due to non-globally Lipschitz conditions.
Ait-Sahalia Type Model
Polynomially growing drift and superlinear diffusion coefficients pose challenges.
Previous methods like Euler-Maruyama not suitable for such models.
Explicit Milstein-Type Schemes
Novel class introduced, focusing on strong convergence with order one.
Corrective mapping Φh incorporated to handle polynomially growing coefficients.
Error Analysis
Lemmas provide moment bounds and continuity properties for solutions.
Error analysis conducted to estimate the error terms in the proposed scheme.
Mean-Square Convergence
Theorem proves expected order-one mean-square convergence for the proposed scheme.
Stats
"The expected order-one mean-square convergence is attained for the proposed scheme."
"The model has a polynomially growing drift that blows up at the origin."