7
$\begingroup$

We know about Kolmogorov Criterion for the tightness of a stochastic process $X_n(t)$

1.The sequence $(X_{n}(0))_{n\geq0}$ is tight.

2.There exist constants $\gamma\geq0$,$\alpha>1$, $K>0$ and an integer $n_0$ such that $$E(|X_{n}(t_{2})-X_{n}(t_{1})|^{\gamma})\leq K|t_{2}-t_{1}|^{\alpha}, \forall n \geq n_0$$ for all $t_{1},t_{2}$.

My first question: what should the $n_0$ depend? Could it depend on the $t_{1}$ and $t_{2}$?

My second question: Is there any other criterion for tightness with the parameter $\alpha=1$ for the version of the moment condition?

$\endgroup$
2
  • $\begingroup$ What space is your stochastic process in? If it's say C[0,1], then the result holds for all n and all t1,t2 in [0,1] $\endgroup$
    – Alex R.
    Sep 17, 2010 at 20:15
  • $\begingroup$ Yes it is in space C[0,1], thanks for your help $\endgroup$
    – syh2010
    Sep 20, 2010 at 6:54

1 Answer 1

2
$\begingroup$

$n_0$ must be independent of $t_1$ and $t_2$, of course. If it's not, the processes might be even discontinuous. For instance, $X_n$ is a Poisson process with parameter $1/n$. Then $$E(|X_n(t_1)-X_n(t_2)|^2)\le |t_1-t_2|^2$$ for all $n>|t_1-t_2|^{-1}$ (for all $n\ge 1$ if $t_1=t_2$).

And the same answer works for the second question: when $\alpha=1$, the processes need not to be continuous. In some special cases, where you have higher moments controlled by a lower one polynomially, it may help (e.g. in the Gaussian case $\gamma=2$ and $\alpha=1$ is enough).

$\endgroup$

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.