Suppose $(X,T)=\{(X_n,T_n)\}_{n\geq0}$ is a Markov renewal process, where $X$ is a finite-state, discrete-time Markov chain with state space $\{1,2,...,R\}$. $T$ is the additive component, more precisely, $T_{n+1}-T_n$ is time the Markov chain $X$ stays in state $X_n$ and may depend on $X_n$ only. Let $\tau_m=\{\min\limits_{n\geq0} n: T_n>m\}$. Does there exist a result that says \begin{equation} \frac{\tau_m-m\mu}{\sqrt{m}\sigma}\underset{m \rightarrow \infty} {\rightarrow}\mathcal{N}[0,1],\end{equation} for some $\mu$ and $\sigma$? If yes, could someone point me in the right direction. Thanks
3
-
$\begingroup$ randomservices.org/random/renewal/LimitTheorems.html $\endgroup$– Carlo BeenakkerCommented Jan 28, 2014 at 13:44
-
$\begingroup$ I don't think theorems for renewal processes automatically apply to Markov renewal processes. $\endgroup$– MthQCommented Jan 28, 2014 at 13:48
-
$\begingroup$ I think my old question mathoverflow.net/questions/100903/… is quite closely related (and would give you the result that you're looking for). $\endgroup$– Anthony QuasCommented Jan 29, 2014 at 5:46
Add a comment
|