Table of Contents
Fetching ...

Optimal Stopping of a Brownian Excursion and an $α$-dimensional Bessel Bridge

David Hobson, Jingfei Liu

TL;DR

The paper addresses optimal stopping for $\alpha$-dimensional Bessel bridges with power payoff $x^n$, unifying the Brownian excursion ($\alpha=3,n=1$) as a key special case. It develops a free-boundary framework with self-similar structures, yielding an explicit solution for the Brownian excursion and a power-series construction for the general $\alpha,n$ case, along with a verification scheme. For the Brownian excursion, the optimal boundary is determined by $C=2e^{-C^{2}/2}\int_{0}^{C}e^{u^{2}/2}\,du$, and the value function is given piecewise by a closed form in the continuation region and by $x$ in the stopping region. In the general setting, the problem reduces to a squared-Bessel representation with a unique root $Z_{\alpha,n}$ of $F_{\alpha,n}(Z)=0$ that defines the boundary, together with a power-series solution for the associated value function; explicit solvable cases (e.g., $n=\alpha-2$ or $\alpha=n$) illustrate the approach. The results extend the literature on optimal stopping for Brownian and Bessel bridges and offer insights applicable to diffusion control and pinning-type financial problems.

Abstract

We study the optimal stopping of an $α$-dimensional Bessel bridge for the payoff $φ(x)=x^n$, where $α,n>0$. As a special case we consider the Brownian excursion with the identity function as the payoff ($α=3,n=1$). For the Brownian excursion we can give an explicit solution but in the general case we provide a complete solution via a power series expansion.

Optimal Stopping of a Brownian Excursion and an $α$-dimensional Bessel Bridge

TL;DR

The paper addresses optimal stopping for -dimensional Bessel bridges with power payoff , unifying the Brownian excursion () as a key special case. It develops a free-boundary framework with self-similar structures, yielding an explicit solution for the Brownian excursion and a power-series construction for the general case, along with a verification scheme. For the Brownian excursion, the optimal boundary is determined by , and the value function is given piecewise by a closed form in the continuation region and by in the stopping region. In the general setting, the problem reduces to a squared-Bessel representation with a unique root of that defines the boundary, together with a power-series solution for the associated value function; explicit solvable cases (e.g., or ) illustrate the approach. The results extend the literature on optimal stopping for Brownian and Bessel bridges and offer insights applicable to diffusion control and pinning-type financial problems.

Abstract

We study the optimal stopping of an -dimensional Bessel bridge for the payoff , where . As a special case we consider the Brownian excursion with the identity function as the payoff (). For the Brownian excursion we can give an explicit solution but in the general case we provide a complete solution via a power series expansion.
Paper Structure (4 sections, 8 theorems, 70 equations, 1 table)

This paper contains 4 sections, 8 theorems, 70 equations, 1 table.

Key Result

Lemma 2.2

Let $f:[0,C] \to [0,\infty)$ be given by eq:fdef with $A=0$ and $B$ given by eq:Bvalue. Then $f(0)=B$ and $f'(0)=0$. Moreover, $f$ is convex with $f(y)\geq y$ for all $y \in [0,C]$.

Theorems & Definitions (18)

  • Definition 2.1
  • Lemma 2.2
  • Theorem 2.3
  • proof
  • Definition 3.1
  • Remark 3.2
  • Corollary 3.3
  • Lemma 3.4
  • proof
  • Proposition 3.5
  • ...and 8 more