Stein's Method for Tempered Stable Distributions

Abstract

In this article, we develop Stein characterization for two-sided tempered stable distribution. Stein characterizations for normal, gamma, Laplace, and variance-gamma distributions already known in the literature follow easily. One can also derive Stein characterizations for more difficult distributions such as the distribution of product of two normal random variables, a difference between two gamma random variables. Using the semigroup approach, we obtain estimates of the solution to Stein equation. Finally, we apply these estimates to obtain error bounds in the Wasserstein-type distance for tempered stable approximation in three well-known problems: comparison between two tempered stable distributions, Laplace approximation of random geometric sums, and six moment theorem for the symmetric variance-gamma approximation of functionals of double Wiener-It$\ddot{\text{o}}$ integrals. We also compare our results with the existing literature.

arXiv comment24 pages

References in corpus (7)

• Tempered stable distributions and processes
• Stein's method and the Laplace distribution
• On Stein's Method for Infinitely Divisible Laws With Finite First Moment
• Stein's method for discrete Gibbs measures
• Exponential approximation and Stein's method of exchangeable pairs
• On Stein's Method for Multivariate Self-Decomposable Laws
• On Discrete Gibbs Measure Approximation to Runs