Embedding laws in diffusions by functions of time

TL;DR

This paper constructs a unique stopping time for Brownian motion to embed a given measure using functions of time, extending to recurrent diffusions, with minimality and optimality properties.

Contribution

It provides a constructive probabilistic proof of the embedding theorem using functions of time, introducing a novel application of the Lévy metric and extending results to recurrent diffusions.

Findings

Existence and uniqueness of embedding functions for Brownian motion.

Minimality of the stopping time in Monroe's sense.

Optimality in truncated expectation among all embeddings.

Abstract

We present a constructive probabilistic proof of the fact that if $B=(B_t)_{t\ge0}$ is standard Brownian motion started at $0$, and $\mu$ is a given probability measure on $\mathbb{R}$ such that $\mu(\{0\})=0$, then there exists a unique left-continuous increasing function $b:(0,\infty)\rightarrow\mathbb{R}\cup\{+\infty\}$ and a unique left-continuous decreasing function $c:(0,\infty)\rightarrow\mathbb{R}\cup\{-\infty\}$ such that $B$ stopped at $\tau_{b,c}=\inf\{t>0\vert B_t\ge b(t)$ or $B_t\le c(t)\}$ has the law $\mu$. The method of proof relies upon weak convergence arguments arising from Helly's selection theorem and makes use of the L\'{e}vy metric which appears to be novel in the context of embedding theorems. We show that $\tau_{b,c}$ is minimal in the sense of Monroe so that the stopped process $B^{\tau_{b,c}}=(B_{t\wedge\tau_{b,c}})_{t\ge0}$ satisfies natural uniform integrability conditions expressed in terms of $\mu$. We also show that $\tau_{b,c}$ has the smallest truncated expectation among all stopping times that embed $\mu$ into $B$. The main results extend from standard Brownian motion to all recurrent diffusion processes on the real line.