0
$\begingroup$

If a sequence of Dirichlet forms convergence to 0, then what about the diffusion processes associated with these Dirichlet forms? Do the finite dimensional distributions of them converges weakly? and what are the limits?

$\endgroup$
4
  • $\begingroup$ What do you mean exactly by "finite dimensional distribution of the diffusion processes" associated with the Dirichlet forms? Perhaps the limit (for $t\to \infty$) of the associated semigroups? $\endgroup$ Jan 27, 2013 at 19:36
  • $\begingroup$ The sequence of Dirichlet forms depend on a parameter, So the convergence of the associated diffusion processes are about this parameter not time $t\rightarrow\infty$. $\endgroup$ Jan 27, 2013 at 22:00
  • $\begingroup$ Yes, this is clear to me. Say, if the forms are $(a_n)_{n\in \mathbb N}$, then I am talking about convergence of the associated semigroups $((e^{-ta_n})_{t\ge 0})_{n\in \mathbb N}$. My question was - and is -: what is the "finite dimensional distribution" of the diffusion process (governed by $(e^{-ta_n})_{t\ge 0}$, say)? $\endgroup$ Jan 27, 2013 at 23:10
  • $\begingroup$ yes, That is what I mean. $\endgroup$ Jan 28, 2013 at 20:38

1 Answer 1

1
$\begingroup$

Kato shows in §VI.3 of his book "Perturbation theory for linear operators" that in particular if a sequence of Dirichlet forms (or, more generally, of bounded closed sesquilinear forms) converges to a limiting form $a$ (in a certain quite natural sense), then the associated operators converge to the operator associated with $a$ in the norm resolvent sense. This in turn implies norm convergence of the associated semigroups, e.g. because of the representation of the semigroups via the backward Euler scheme applied to the resolvents. I guess this answers your question about the diffusion processes associated with the forms.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.