Brownian Skorokhod Embedding¶
Represent a prescribed probability law as Brownian motion observed at an adapted stopping time, with admissibility conditions stated separately.
Core Idea¶
A Brownian Skorokhod embedding chooses a stopping time \(T\) adapted to Brownian motion \(B\) so that \(B_T\) has a prescribed probability law \(\mu\). It is an equality in distribution, not a claim that an original random variable has been placed injectively or pathwise into Brownian motion. For a centered finite-variance law, classical constructions can also control expected stopping time; Billingsley's one-law theorem gives a stop with \(E[T]=\int x^2\,\mu(dx)\). That property belongs to an admissible construction, not to every stop satisfying \(B_T\sim\mu\).[ref-108665861969][ref-a914acbf2b87]
The frozen title “Skorokhod's embedding theorem” names either or both of a one-law result and an iterated IID partial-sum result. This draft takes the stable Brownian embedding problem and method as its identity. The two theorem-level senses remain unresolved lineage, not aliases automatically covered by this entry.[^ref-108665861969]
Scope of Application¶
The Brownian carrier, its filtration, an adapted stop, and the target-law match are constitutive. Minimality, integrability, or a path-functional extremum may be needed for a particular proof or application, but is not automatic. A noncentered target, a target lacking the relevant moments, or a general diffusion requires a separately stated formulation. A law-matching stop can even have infinite expected time.[^ref-67c2fb0f0c35]
In probability, repeated stops can represent centered IID partial sums in joint law; control of the random clock then helps transfer Brownian limit arguments. In mathematical finance under specified continuous-martingale and market assumptions, compatible stopped-Brownian constructions can share a terminal marginal while yielding different path-dependent payoffs. Neither setting makes the embedding unique.[ref-108665861969][ref-49ac9b16f307]
Clarity¶
The target law fixes the stopped value's distribution, not the preceding path or the stopping time. For the symmetric \(\{-1,1\}\) law, first exit of Brownian motion from \((-1,1)\) is one valid stop, but other laws and other admissibility goals call for different constructions. This is distinct from generic optional stopping and from Skorokhod's representation theorem for almost-sure coupling of weakly convergent laws.[ref-67c2fb0f0c35][ref-a914acbf2b87]
Live Embedding requires a faithful injective structural placement, so it is a lexical neighbor rather than a strict DAG parent. The proposed strict prerequisite is live Stopping time: without a nonanticipating random observation time, the scoped Brownian embedding cannot be formed.
Manages Complexity¶
The method moves a target distribution into a continuous Brownian setting where path and scaling tools are available. The resulting random clock makes the cost of that move visible. Repeated embeddings may simplify a random-walk limit argument, but matching the stopped values does not by itself preserve fixed-time laws, maxima, or path-dependent prices.[ref-49ac9b16f307][ref-a914acbf2b87]
Abstract Reasoning¶
Specify the target law, Brownian start and filtration. Construct an adapted \(T\) and prove \(B_T\sim\mu\). Only then check the particular integrability, minimality, joint-law, or payoff property the task needs. For a Rademacher target, first exit from \((-1,1)\) gives the desired symmetric law by continuity and symmetry; repeating the relative exit rule after each stop yields IID stopped increments by the strong Markov property.[ref-108665861969][ref-67c2fb0f0c35]
Billingsley's separate iterated theorem concerns increasing stops whose Brownian values match IID partial sums jointly. Its construction-specific moment estimates must not be imputed to every law-matching stop.[^ref-108665861969]
Knowledge Transfer¶
The same roles map to unlike settings. In random-walk coupling, the target is an increment law, repeated adapted stops provide joint partial-sum laws, and accumulated stop durations matter. In robust pricing, a terminal marginal is the target after the framework's transformation, admissible Brownian constructions preserve it, and a specified path payoff varies across constructions. The terminal-law match transfers literally; Brownian filtration and admissibility assumptions must be rechecked rather than imported by analogy.[ref-49ac9b16f307][ref-a914acbf2b87]
[^ref-108665861969]: Patrick Billingsley, Probability and Measure, 3rd ed., Wiley (1995), §37, Theorems 37.6–37.7, pp.519–521. University-hosted copy was search-index accessible, but direct PDF open was intermittent. [^ref-a914acbf2b87]: Jan Obłój, “The Skorokhod Embedding Problem and Its Offspring”, Probability Surveys 1 (2004), 321–392, especially §§2–3 and 11.2. [^ref-67c2fb0f0c35]: Jan Obłój, The Skorokhod Embedding Problem, Chapter 4 lecture notes, pp.38–40. [^ref-49ac9b16f307]: David Hobson, The Skorokhod Embedding Problem and Model-Independent Bounds for Option Prices, original author lecture notes (2009), pp.1–3, §4.1 and §5.
Relationships to Other Abstractions¶
Current abstraction Brownian Skorokhod Embedding Domain-specific
Parents (1) — more general patterns this builds on
-
Brownian Skorokhod Embedding presupposes Stopping time Domain-specific
A Brownian Skorokhod embedding requires an adapted stopping time to select the distribution-matching Brownian value.
Hierarchy path (1) — routes to 1 parentless root
- Brownian Skorokhod Embedding → Stopping time → Threshold
Neighborhood in Abstraction Space¶
Brownian Skorokhod Embedding sits in a sparse region of the domain-specific corpus (63rd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.
Family — Foundations of Probability & Inference (29 abstractions)
Nearest neighbors
- Residence Time (Statistics) — 0.86
- Cramér's Theorem (Large Deviations) — 0.85
- Matrix Analytic Method — 0.85
- Stein's Unbiased Risk Estimate — 0.84
- Gambling and information theory — 0.84
Computed from structural-signature embeddings · 2026-10-08