Interpret as a conformal bijection onto . If it is merely an injective map into a larger target, the correct exit domain is instead. Let be the exit time from , write , and identify planar Brownian motion with . The Cauchy-Riemann equations give , , and both are harmonic functions. The Itô formula, localized on compact subsets of , yields
Thus both coordinates are continuous local martingales, with
Since never vanishes, is strictly increasing before . Let , defined for . The optional time-change theorem gives continuous local martingale coordinates for , with bracket matrix . The Lévy characterization of multidimensional Brownian motion makes a standard planar Brownian motion started at , up to its lifetime. Therefore conformal invariance of planar Brownian motion takes the form
To identify the lifetime, exhaust by relatively compact open sets with nested closures. The images exhaust because is a homeomorphism onto . Each stopped transformed path exits at precisely the clock value of its exit from . Increasing these exit times gives the exit lifetime of Brownian motion from , including a possibly infinite lifetime. This identifies with that lifetime and needs no global extension of to the boundary.
The Lévy characterization of Brownian motion states that an adapted continuous real process with is standard Brownian motion relative to if and only if it is a continuous local martingale and . The multidimensional version replaces the bracket identity by .
For the converse direction, a standard Brownian motion is a continuous martingale, and its independent centered increments give quadratic variation . Distinct independent coordinates have zero quadratic covariation.
For the substantive direction, fix a vector . The Itô formula applied to
shows that its drift is zero: the second-order term is and cancels the time derivative. Thus is a complex local martingale. On its modulus is the deterministic value ; localization and the dominated convergence theorem make it a true martingale. Hence
This conditional characteristic function is the deterministic characteristic function of . For , multiplying by and taking expectations, then using uniqueness of characteristic functions, proves that has this multivariate normal distribution and is independent of . Iterated conditioning gives independent increments with Gaussian distributions. Continuity, the zero initial value and these increment laws are the definition of standard Brownian motion. The case proves the requested statement; the vector form will also be used below.
Use two standard results: for every continuous local martingale , there is a unique continuous adapted increasing quadratic variation , starting at zero, such that is a local martingale; and a continuous finite-variation local martingale is constant.
Define the quadratic covariation by the polarization identity
This is continuous, adapted, starts at zero, and is of finite variation on compact time intervals. Since , subtraction of the two quadratic-variation identities shows that is a continuous local martingale. In the zero-start convention this is exactly .
If is another continuous adapted finite-variation process with and the same local martingale property, then is both a continuous local martingale and of finite variation. It is constant, and its initial value is zero, so it vanishes. Uniqueness is up to indistinguishability of stochastic processes.
The normalization is essential. If arbitrary nonzero initial constants are allowed in the martingale class and this normalization is omitted, adding a constant to the bracket preserves the claimed property. The usual definition of quadratic covariation always includes the zero initial value.
For a continuous vector semimartingale and , the Itô formula is
The Itô product rule, or stochastic integration by parts, is
The additional second-order term is the quadratic covariation.
For general semimartingales with càdlàg paths, use left limits in the integrands. The Itô formula for semimartingales with jumps, for time-independent , is
Here denotes the continuous local martingale part; the continuous bracket is that of these parts. For time-dependent add and insert in the other evaluations of . The integration by parts formula becomes
For continuous paths the jump sum vanishes, recovering the first two formulas.
A continuous martingale is an adapted process with almost surely continuous paths, for every , and
A continuous local martingale is continuous and adapted and has a localizing sequence of stopping times almost surely such that every is a true martingale. The initial value is normally integrable; the frequently used class is normalized to start at zero.
The usual conditions for a filtration are completeness and right continuity. Completeness means that contains every subset of each -null set in the ambient ; right continuity means . These conditions concern the filtered probability space, whereas continuity of a martingale concerns its sample paths.
The Girsanov theorem construction proves a weak stochastic solution, but does not by itself establish a strong stochastic solution. It constructs first, and then constructs a new driving Brownian motion . Adaptation of to the original filtration of does not by itself prove adaptation to the smaller natural filtration of . Inverting this relation requires an additional result; uniqueness in law alone is insufficient in general.
For the particular coefficients in part (b), however, the literal claim that a strong solution may fail is incorrect. The strong existence theorem for additive-noise SDEs with bounded measurable drift gives strong existence and pathwise uniqueness for bounded Borel state-dependent and unit diffusion. Consequently every weak realization here is determined by its driving noise and initial value. This theorem is deeper than the Girsanov theorem argument and no proof is needed here. Examples with path-dependent drift or with a nonconstant diffusion coefficient do not contradict it.
The valid distinction is therefore: the argument in part (b) has established weak existence and uniqueness in law; strong existence requires an additional theorem, and in this class that theorem is available. A primary reference for the correction is On strong solutions and explicit formulas for solutions of stochastic integral equations.
Fix and a finite horizon . Begin under Wiener measure with , and let bound . The bounded progressively measurable process is an admissible Itô integral integrand. Put
The Novikov condition follows from . With , the Girsanov theorem says that is Brownian motion. Therefore a weak stochastic solution exists, since
Consistent finite-horizon densities on canonical continuous path space give a solution for all finite times. No uniform integrability at infinite time is needed for this extension.
For uniqueness in law, take any weak stochastic solution under a measure and reverse the change of measure:
Again the Novikov condition holds. Under , is Brownian motion, so the law of is the fixed shifted Wiener measure. Substitution of gives
Under shifted Wiener measure, the stochastic integral on the right has a fixed measurable version as a functional of the canonical path. It can be defined by simple predictable processes converging in the integrand norm and an almost surely convergent subsequence of their integrals. Equivalence of the two finite-horizon measures preserves that version. Thus for every bounded measurable path functional ,
The right-hand side depends only on , not on the chosen solution. This proves uniqueness in law on every finite horizon, and hence on the whole continuous path space. This is Weak existence and uniqueness in law for an additive-noise SDE with bounded drift. For merely measurable , one should not assume that naive deterministic left-endpoint samples of converge; the predictable approximation is the appropriate definition.
A weak solution of a stochastic differential equation consists of a filtered probability space, an adapted continuous process , and a Brownian motion relative to that filtration, with specified initial distribution, such that for every finite the coefficient integrals are finite and
In one dimension the integrability conditions are and . The noise integral is the Itô integral. A solution on a prescribed stochastic basis with prescribed is distinguished from this weak construction.
A strong solution of a stochastic differential equation uses the prescribed and initial variable, and is adapted to their augmented natural filtration; thus no additional randomness is needed. For deterministic , each is determined by the driving Brownian motion through time . Merely being adapted to a larger filtration is insufficient for this definition.
Uniqueness in law means that any two weak stochastic solutions with the same initial distribution have the same law of on continuous path space. Pathwise uniqueness means that two solutions on one stochastic basis, with the same and almost surely equal initial values, are indistinguishable stochastic processes. The latter compares paths under the same noise, rather than just comparing their probability distributions.
The reference Brownian motion must start at , and both laws must use the same stopped-path rule. Under the reference probability measure let , let be its exit time from , and put and . The Girsanov density for a stopped Bessel process, as a functional of the stopped reference path, is
Use the predictable process , with any harmless choice at the endpoint. On , . The Novikov condition holds; under the density , the Girsanov theorem makes a Brownian motion. Before exit, , which is the required stochastic differential equation. The drift is a locally Lipschitz function on , so pathwise uniqueness and uniqueness in law hold up to this exit time.
For a version involving just the endpoint and an ordinary integral, the Itô formula for gives
Thus
The stopped density is measurable from the stopped path, so it is also the Radon-Nikodym derivative of the stopped-path laws. The stopping times in this formula are evaluated on the reference path, rather than taken from a different realization of . A zero-start reference Brownian motion would have a different deterministic initial value and could not dominate this law.
The Novikov condition holds on every horizon, since
Therefore is a true martingale and the Girsanov theorem applies through each finite .
The deterministic bound also gives a genuine infinite-horizon density. For any , the stochastic exponential identity is
A nonnegative local martingale is a supermartingale, so
This bound gives uniform integrability. Also is a L2-bounded continuous martingale, since localization and the Itô isometry give . The L2 martingale convergence theorem gives a finite limit ; hence and . Weighting by defines an equivalent probability measure on . This is the bounded-bracket criterion for a stochastic exponential.
For a continuous local martingale , its stochastic exponential is
The Itô formula gives and , so is a strictly positive local martingale. The finite-horizon Girsanov theorem says: if is a true martingale on and on , then for every continuous local martingale under ,
For an infinite-horizon change of measure, assume is a uniformly integrable martingale and use its terminal density . In particular, if , the process is Brownian motion under the new probability measure, by the Lévy characterization of Brownian motion.
The quadratic variation cannot change. Adding the continuous finite-variation process leaves . Also the same pathwise partition sums defining have the same limit under an absolutely continuous change of measure; convergence in probability transfers by integrability of the density. Thus the bracket of the unchanged process , now a semimartingale, is also unchanged. The local martingale property itself need not survive. This is quadratic covariation under an absolutely continuous measure change.
On the same measurable space, absolute continuity of measures means
The Radon-Nikodym theorem then supplies a nonnegative measurable density with and . The converse follows directly from this formula. Strict positivity of the Radon-Nikodym derivative gives equivalence of the two probability measures; absolute continuity of measures alone need not give equivalence.
For the simple predictable process from part (a), define its stochastic integral by
The time-zero value contributes nothing. Write . If and , then is -measurable and . If , then . Boundedness ensures integrability, so all off-diagonal terms vanish by conditional expectation. Consequently
The quadratic variation theorem says is a local martingale. A bounded continuous martingale is square-integrable, and localization gives the conditional second-moment identity
Multiplication by and summation therefore give the Itô isometry
For Brownian motion, . More generally the Itô isometry holds for every predictable process with finite right-hand side, against a continuous square-integrable martingale. Indeed, on a fixed horizon the quadratic-variation measure is finite. Part (c) approximates the integrand by simple predictable processes; the displayed equality makes the elementary integrals Cauchy in and defines their unique limit. The finite sum is exactly the elementary case of the Itô isometry.
Let be the closure of the simple predictable processes in the Hilbert space . Common refinement of deterministic partitions makes these processes a vector space. To prove density of simple predictable processes for finite measures, suppose . Since is a finite measure, the Cauchy-Schwarz inequality gives , so
defines a finite signed measure. The indicator of each predictable generating rectangle, including each time-zero rectangle, is a simple predictable process; hence vanishes on these rectangles.
Finite intersections of the rectangles are again rectangles or empty. Add the whole space to this pi-system. Its -mass is zero because is simple and converges to in by the dominated convergence theorem. The zero sets of form a lambda-system now that its total mass is zero. The Pi-lambda theorem therefore gives for every . Taking the sets where and gives -almost everywhere.
Thus and the orthogonal complement criterion for a closed subspace of a Hilbert space yields
Omitting the time-zero term would make this conclusion false when is concentrated at time zero.
For fixed , define the left-limit sigma-algebra by . The section at of a generating rectangle of the predictable sigma-algebra is when , and empty otherwise. In the first case . Sections of the time-zero generators are empty for .
The collection of subsets of sample-time space whose section at belongs to is a sigma-algebra: sections commute with complements and countable unions. It therefore contains all of . Apply this to for every Borel set . Thus predictable sections at deterministic times satisfy
At the same argument gives -measurability, using the convention . The PDF asks for information strictly before ; the TeX aid drops the minus sign.
Use a filtered probability space , and put . The predictable sigma-algebra on is
Equivalently it is generated by left-continuous adapted processes. A real-valued previsible process is a function that is measurable for and the Borel sigma-algebra of .
A simple predictable process has the form
where is bounded and -measurable, and each is bounded and -measurable. Set after ; allowing a final infinite interval gives an equivalent dense class. The values may be restricted to finitely many values if desired, by approximating bounded measurable functions by simple functions. The initial-time term matters for arbitrary finite measures on : a measure can charge time zero, even though the stochastic integral against a continuous martingale does not.
The printed uniqueness claim omits nondegeneracy of the limit. The zero process is a Lévy process, and every yields that limit. We first classify all possibilities under the stated assumptions, then identify the unique nonzero one.
For , part (b) gives
for each fixed . If , the characteristic functions tend to zero for every nonzero parameter and remain one at zero. This discontinuous function cannot be the characteristic function of a probability law. The characteristic-function convergence theorem therefore rules out even one-dimensional weak convergence of random variables at a positive time.
For joint convergence, take and write . Independent increments give
At , this tends to . These are exactly the joint laws of standard Brownian motion: increments are independent centred normals with variances . The assumed limiting Lévy process thus has the Brownian law.
For , the same joint characteristic functions tend to one. Every limiting vector is zero. The càdlàg path convention then makes the limiting process identically zero almost surely: first intersect the zero-value events at rational times, then use right continuity. The complete scaling classification of a continuous-time symmetric simple random walk is
In particular, if a nonzero limiting Lévy process is intended, the unique exponent is , and
The finite-dimensional Brownian law, together with the Lévy càdlàg convention, determines the process law and its continuous version. No assertion of path-space convergence of the approximating walks is needed here.
Both Poisson processes start at zero and have càdlàg paths. On disjoint time intervals their increments are independent, and independence of the two processes makes the difference increments independent as well. Their increment laws depend only on interval length, so has stationary increments. For a time interval of length , the probability that either Poisson process has a jump is . This proves stochastic continuity. Consequently the difference is a Lévy process, equivalently a continuous-time symmetric simple random walk with total jump rate one and jumps and each chosen with probability .
The characteristic function of a Poisson variable of mean is . Independence therefore gives
Thus, in the negative-exponent convention of part (a),
In the positive-exponent convention the answer is . The Compound Poisson process has mean zero and variance , consistent with the expansion .
A real Lévy process is a stochastic process with almost surely, independent increments, stationary increments, and stochastic continuity. The latter means in probability as . Its usual path convention is càdlàg: paths are right-continuous and have left limits. Equivalently one starts with the increment and continuity-in-probability conditions and then chooses the càdlàg version.
The Lévy characteristic exponent is specified here by
Some conventions use instead, with . The sign convention must be fixed before computing the exponent. The zero process satisfies this definition; excluding it requires an additional nondegeneracy condition.
Fix and start at with . Write the independent Brownian coordinates as and . Let
On the path has stayed in the upper strip and first exits the cross-shaped domain at its right or left side. The boundary value is or , respectively. Let if and if . The Brownian exit from an interval probabilities give
Couple all heights using the same two Brownian paths. By part (a), almost surely, and continuity makes finite. Thus
Let . On the complement of this exceptional event, ; on the event itself, . The representation in part (c) therefore yields
This proves the harmonic limit along an infinite strip arm:
Far up the arm, the bottom is reached with vanishing probability before the horizontal coordinate exits. The limit is consequently determined by the two side values alone.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact