“…We will establish our results in some generality, at the level of Markov additive processes in general state space, as studied in a large deviations context by Ney and Nummelin (1987a, b), de Acosta (1988), de Acosta and Ney (1998, and references therein, following along the lines of the seminal papers of Donsker and Varadhan (1975, 1976, 1983. Thus, our results differ from known importance sampling results given e.g.…”
Section: Introductionmentioning
confidence: 63%
“…where π is the stationary measure of the Markov chain {X n }; this is true, for example, if (R) holds and 0 ∈ dom Λ P , or alternatively if (M ) holds and the set dom ψ is open [Ney and Nummelin (1987a), Lemma 3.3 and Lemma 5.2]. From (3.6), we then see that the central tendency of S 1 , S 2 , .…”
Section: Notation Hypotheses and Estimation Resultsmentioning
confidence: 94%
“…For the proofs, see Ney and Nummelin (1987a), Sections 3 and 4, and Iscoe, Ney and Nummelin (1985), Lemma 3.1.…”
Section: Nonnegative Kernels Eigenvalues and Eigenvectorsmentioning
Let {(X n , S n ) : n = 0, 1, . . .} be a Markov additive process, where {X n } is a Markov chain on a general state space and S n is an additive component on R d . We consider P {S n ∈ A/ , some n} as → 0, where A ⊂ R d is open and the mean drift of {S n } is away from A. Our main objective is to study the simulation of P {S n ∈ A/ , some n} using the Monte Carlo technique of importance sampling. If the set A is convex, then we establish: (i) the precise dependence (as → 0) of the estimator variance on the choice of the simulation distribution; (ii) the existence of a unique simulation distribution which is efficient and optimal in the asymptotic sense of Siegmund (1976). We then extend our techniques to the case where A is not convex. Our results lead to positive conclusions which complement the multidimensional counterexamples of Glasserman and Wang (1997).
“…We will establish our results in some generality, at the level of Markov additive processes in general state space, as studied in a large deviations context by Ney and Nummelin (1987a, b), de Acosta (1988), de Acosta and Ney (1998, and references therein, following along the lines of the seminal papers of Donsker and Varadhan (1975, 1976, 1983. Thus, our results differ from known importance sampling results given e.g.…”
Section: Introductionmentioning
confidence: 63%
“…where π is the stationary measure of the Markov chain {X n }; this is true, for example, if (R) holds and 0 ∈ dom Λ P , or alternatively if (M ) holds and the set dom ψ is open [Ney and Nummelin (1987a), Lemma 3.3 and Lemma 5.2]. From (3.6), we then see that the central tendency of S 1 , S 2 , .…”
Section: Notation Hypotheses and Estimation Resultsmentioning
confidence: 94%
“…For the proofs, see Ney and Nummelin (1987a), Sections 3 and 4, and Iscoe, Ney and Nummelin (1985), Lemma 3.1.…”
Section: Nonnegative Kernels Eigenvalues and Eigenvectorsmentioning
Let {(X n , S n ) : n = 0, 1, . . .} be a Markov additive process, where {X n } is a Markov chain on a general state space and S n is an additive component on R d . We consider P {S n ∈ A/ , some n} as → 0, where A ⊂ R d is open and the mean drift of {S n } is away from A. Our main objective is to study the simulation of P {S n ∈ A/ , some n} using the Monte Carlo technique of importance sampling. If the set A is convex, then we establish: (i) the precise dependence (as → 0) of the estimator variance on the choice of the simulation distribution; (ii) the existence of a unique simulation distribution which is efficient and optimal in the asymptotic sense of Siegmund (1976). We then extend our techniques to the case where A is not convex. Our results lead to positive conclusions which complement the multidimensional counterexamples of Glasserman and Wang (1997).
“…In the literature, the following stronger condition is often used (see, e.g. [26,27]), but we only need it for some special cases in Section 5.…”
Section: Markov Additive Process and The Wiener-hopf Factorizationmentioning
confidence: 99%
“…If S X is countable, then (3b) is automatically satisfied under assumption (3a). For the general case, useful sufficient conditions are given in Proposition 3.1 of [26].…”
Section: Markov Additive Process and The Wiener-hopf Factorizationmentioning
We extend the framework of Neuts' matrix analytic approach to a reflected process generated by a discrete time multidimensional Markov additive process. This Markov additive process has a general background state space and a real vector valued additive component, and generates a multidimensional reflected process. Our major interest is to derive a closed form formula for the stationary distribution of this reflected process. To this end, we introduce a real valued level, and derive new versions of the Wiener-Hopf factorization for the Markov additive process with the multidimensional additive component. In particular, it is represented by moment generating functions, and we consider the domain for it to be valid.Our framework is general enough to include multi-server queues and/or queueing networks as well as non-linear time series which are currently popular in financial and actuarial mathematics. Our results yield structural results for such models. As an illustration, we apply our results to extend existing results on the tail behavior of reflected processes.A major theme of this work is to connect recent work on matrix analytic methods to classical probabilistic studies on Markov additive processes. Indeed, using purely probabilistic methods such as censoring, duality, level crossing and time-reversal (which are known in the matrix analytic methods community but date back to Arjas & Speed [2] and Pitman [29]), we extend and unify existing results in both areas.
In some cases it is interesting to have inequalities between asymptotic rates. Here, we consider asymptotic rates in the fashion of large deviations. We present inequalities between large deviation rate functions for telegrapher processes in the first part, and inequalities between Lundberg parameters for Markov-additive processes in the second part
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.