The integrated friction is exactlyFor every finite , the roots of have negative real parts. The Green-Kubo formula for diffusion therefore gives the diffusion coefficient directly:ConsequentlyIntegrating the limiting velocity autocorrelation function gives the same value, . Computing before taking the limit also avoids having to justify an interchange of the infinite-time integral and the parameter limit.
Write , so the friction memory kernel is and has area one for every . Its Laplace transform is . The generalized fluctuation-dissipation relation therefore givesTo verify the pointwise time-domain limit, rather than relying only on Laplace transform convergence, consider . The two rootsare negative and satisfy , . Since and , the inverse Laplace transform isThe coefficient of the slow exponential tends to , and that of the fast exponential tends to zero. HenceThis is the velocity autocorrelation function of an Ornstein-Uhlenbeck process satisfying . The friction memory kernel becomes local in time. Consistently, the even noise autocorrelation tends to in the sense of distributions, fixing the white noise normalization. The finite- right derivative does not converge to the right derivative of the limiting function: there is a short initial relaxation layer.
Since , its supremum is nonnegative. For , continuity givesThe events increase with . Part (iii) therefore yields . Letting gives ; letting gives finiteness almost surely. ConsequentlyIts probability density function is , with no atom at zero. Using strict exceedances above also avoids presupposing that the infinite-horizon supremum is attained.
Let and . Set . The exit time from is at most the exit time from , so part (ii) gives . Continuity ensures that exactly one of the two different boundary levels is reached at .
The stopped local martingale lies in , hence is a uniformly integrable martingale. The optional sampling theorem, justified at the terminal time also by bounded convergence theorem, givesThus the boundary hitting probability from a diffusion scale function isFor integer , the events increase to . Indeed, on any path with finite , continuity makes the path bounded below on , so some lower barrier has not yet been hit. The reverse inclusion is immediate. Taking the limit in the displayed probability gives
The stopped representation from part (a)(ii) and the Itô isometry give the exact second-moment identityThe cross term with the initial value is zero because the stopped stochastic integral has zero mean. This is the useful finite-time computation; evaluating the remaining occupation integral explicitly is unnecessary for proving integrability of .
For ,Combining the second-moment identity with part (i) yieldsLet and use monotone convergence:In particular almost surely, a conclusion obtained without assuming finite exit in the stopping argument.
One can also evaluate the limiting second moment exactly. At exit is either or , soBy bounded convergence theorem and part (i), . ThereforeThe same bound also holds at every finite time, by applying to and taking expected values. It improves the preceding bound to if desired.
Take , as needed for a nontrivial interval. Since is continuous and adapted, for every level and ,The infimum is attained on this compact interval and can be computed on the countable dense set . It is therefore -measurable. Thus is a stopping time, and so is , because .
By continuity, , even before knowing that is finite. HenceStopping preserves the local martingale property. The bounded local martingale criterion now shows that is a uniformly integrable martingale; its deterministic bound also gives square integrability. In particular,
Apply the Itô formula to . Since and , the two drift terms cancel:The integrand is continuous and hence locally square integrable after stopping at bounded levels of . Thus is a positive continuous local martingale, with and quadratic variationThe cancellation is the reason is the relevant scale function for this diffusion.
Write the diffusion and drift coefficients asBoth are continuously differentiable, withThese derivatives are bounded. For example bounds both absolute values by one. The mean value theorem therefore makes both coefficients globally Lipschitz. They are also bounded, so satisfy the linear growth condition for an SDE.
The global existence theorem for stochastic differential equations with Lipschitz coefficients consequently givesIn particular, there is no finite explosion time. The solution has continuous paths, so all subsequent level-hitting and stopping arguments apply.
Put . The change of measure and part (a)(iii) yieldFor a centered normal random variable of variance , a Gaussian integral gives the Gaussian quadratic exponential momentHere and for finite positive , so the required positivity holds, including when . ThusExpanding the two exponentials in terms of hyperbolic functions givesAt this is one. As , the denominator tends to , recovering part (a)(i). At it gives the Laplace transform of the integrated square of Brownian motion. The joint terminal-and-integral transform is the Joint Laplace transform of Brownian square and its time integral.
Fix . Part (b)(i) makes a probability measure on , equivalent to since . Its density process on is . The Girsanov theorem, applied to the stochastic exponential with integrand , says thatis standard Brownian motion under . The required local square-integrability follows from path continuity, and the true martingale property was proved in part (b)(i).
By part (b)(ii), under the process is therefore the zero-started Ornstein-Uhlenbeck process:This identifies its full law on continuous path space, rather than just its one-time marginal. It is a centered Gaussian process, since it consists of deterministic Itô integrals against Brownian motion. For , the Itô isometry gives its covariance function:In particular, its Ornstein-Uhlenbeck transition distribution at the terminal time is
The relation between and is the linear equation with . Multiplication by the deterministic integrating factor and the Itô product rule giveHenceThe second equality follows by semimartingale integration by parts. It expresses as an explicit function of the continuous path of , using only an ordinary time integral. This inversion holds before changing measure; is already a continuous semimartingale under the original measure.
The expression obtained in part (a)(iii) gives a direct deterministic bound. For every fixed and ,Thus the bounded local martingale criterion makes a true martingale on every finite time interval. More explicitly, if is a localizing sequence, the identities pass to the limit by dominated convergence, using the displayed bound for . ThereforeThis establishes exactly the finite-horizon property needed for the subsequent change of measure.
The Itô formula for yields . Substituting into the expression for gives the pathwise identityConsequentlyTaking expected values provesThe random variable on the preceding line is bounded by one for , so its integrability is immediate.
Let . Its quadratic variation is . The Itô formula for givesChooseThe chosen process is continuous and adapted, hence progressively measurable. Its time integral is finite on compact intervals because Brownian paths are continuous. The resulting is the stochastic exponential of , and is a positive local martingale.
For , has the normal distribution . Evaluate its Gaussian integral:At the defining random variable is one, consistent with the same formula. Thus
A sufficient condition is that is globally Lipschitz. A more familiar, stronger condition implying this is that is globally Lipschitz and . Both conditions repair the existence issue in part (i).
To connect this to the suggested stochastic differential equation, apply the coordinate and coordinate-product cutoff tests to any solution of the martingale problem. As before they giveDefine, using stochastic integrals against the coordinates of ,These integrals exist even without a global positive lower bound: their quadratic variations are exactly . Indeed,The Lévy characterization of multidimensional Brownian motion identifies as standard -dimensional Brownian motion. By associativity of stochastic integration,The diffusion matrix is globally Lipschitz under the stated condition, and boundedness of gives bounded growth. The global existence theorem for stochastic differential equations with Lipschitz coefficients gives a unique strong solution of a stochastic differential equation and uniqueness in law. Every solution to the martingale problem has just been represented as a weak solution of a stochastic differential equation of this form, so they all have the same law. Conversely, the Itô formula applied to any gives a true martingale, since its stochastic integrand is bounded. Thus the martingale problem is well posed under this sufficient condition.
The existence assertion is false under the hypotheses printed in the source. No measurability of is specified. Even requiring to be a Borel measurable function does not suffice: a coefficient can be positive at every point while approaching zero too rapidly near the starting point. This is the obstruction described by pointwise positivity does not ensure weak existence.
Here is a counterexample in dimension one, with :Suppose that a solution to the martingale problem with domain existed. The cutoff argument from part (a)(ii), now with , makes a continuous local martingale withThe criterion for a continuous local martingale with bounded quadratic variation makes it a square-integrable martingale, and the Itô isometry gives .
Apply the occupation-times formula with the indicator of the singleton . Since that singleton has zero Lebesgue measure,Thus spends zero ordinary time at zero. Away from zero, , soThe Gronwall inequality forces for every . Continuity then forces almost surely, contradicting either the zero occupation time just proved or for that constant path. Consequently
For completeness, the intended construction works if is a Borel measurable function and bounded away from zero on every compact set. In particular, it works for continuous strictly positive . Let be standard -dimensional Brownian motion, and define a clock and its inverse byThe Brownian path on each finite interval has compact range, so the added hypothesis makes finite. The clock is continuous and strictly increasing, and , so . Its inverse is continuous and satisfies . Each is a stopping time for the Brownian filtration.
For , the optional sampling theorem applied at these bounded random times shows thatis a true martingale in the filtration . The inverse-clock substitution , valid almost everywhere, rewrites it asThis proves the Brownian time change for a variable diffusion speed:The factor in the clock is necessary because standard Brownian motion has infinitesimal generator .
TakePart (i) proves that Brownian motion solves this martingale problem. For the converse, let be any continuous adapted process satisfying this martingale problem and .
Set . Choose a smooth cutoff function equal to one on a neighbourhood of the closed ball of radius . Apply the martingale problem to and , and stop at . Inside that ball, their Laplace operators are respectively and . The optional sampling theorem therefore shows thatare martingales. Continuity makes almost surely. Hence every is a continuous local martingale. The Itô product rule and the fact that a continuous finite-variation local martingale is constant identify its quadratic covariations:Here is the Kronecker delta.
The Lévy characterization of multidimensional Brownian motion now applies. To see explicitly why it identifies the entire law, for the Itô formula makesa complex local martingale. Its absolute value on is at most , so the bounded local martingale criterion makes it a true martingale there. For ,This conditional characteristic function identifies as having a centered multivariate normal distribution with covariance matrix , independent of . Together with continuity and , these are exactly the defining properties of standard -dimensional Brownian motion. Thus the martingale problem for Brownian motion has a unique law.
A convenient sufficient condition is that has a bounded gradient. The Itô formula for standard -dimensional Brownian motion givesIf , the Itô isometry givesThus the right side is a square-integrable martingale, rather than merely a local martingale. In particular,There are many nonconstant examples, such as or . More generally it is enough that for each finite . Smooth test functions with compact support satisfy the bounded-gradient condition and will provide the domain in the next part.
A second-order nondivergence-form elliptic operator on a domain has the formOnly the symmetric part of affects the Hessian matrix term, so one may take symmetric. The operator is a uniformly elliptic operator if there is a constant , independent of , such thatA common bounded-coefficient convention additionally requires an upper bound with . The lower bound says that the principal quadratic form cannot degenerate in any direction or at any location.
For an explicit stochastic example, standard -dimensional Brownian motion has infinitesimal generatorIts principal quadratic form is , so it is uniformly elliptic with . More generally, a sufficiently regular stochastic differential equation has diffusion generator ; it is uniformly elliptic when has a uniformly positive lower bound. A conservative diffusion process has no zeroth-order term, since its infinitesimal generator annihilates constants.
There are unlisted articles, also show them or only show them.