Semi-Markov processes and α-invariant distributions

A semi-Markov process is easily made Markov by adding some auxiliary random variables. This paper discusses the I-type quasi-stationary distributions of such “extended” processes, and the α-invariant distributions for the corresponding Markov transition probabilities; and we show that there is an intimate relation between the two. The results have relevance in the study of the time to “absorption” or “death” of semi-Markov processes. The particular case of a terminating renewal process is studied as an example.     

Markov chains with replacement

The following modification of a general state space discrete-time Markov chain is considered: certain transitions are supposed “forbidden” and the chain evolves until there is such a transition. At this instant the value of the chain is “replaced” according to a given rule, and, starting from the new value, the chain evolves normally until there is a forbidden transition again; the cycle is then repeated. The relationship of this modified process to the original one is studied in general terms, with particular emphasis being given to invariant measures. Examples are given which illustrate the results obtained.     

A note on the exponentiality of total hazards before failure

It is well known that a univariate counting process with a given intensity function becomes Poisson, with unit parameter, if the original time parameter is replaced by the integrated intensity. P. A. Meyer (in Martingales (H. Dinges, Ed.), pp. 32–37. Lecture Notes in Mathematics, Vol. 190, Springer-Verlag, Berlin) showed that a similar result holds for multivariate counting processes which have continuous compensators. Even more is true in the multivariate case: If each coordinate process is transformed individually according to a convenient time change, the resulting Poisson processes become independent. Our aim is to show that the continuity assumption of the compensators can be relaxed and, when the jumps of the compensator become small, we obtain the independent Poisson processes as a limit. An application for testing goodness-of-fit in survival analysis is given.     

Bayesian spatial modeling of genetic population structure

Natural populations of living organisms often have complex histories consisting of phases of expansion and decline, and the migratory patterns within them may fluctuate over space and time. When parts of a population become relatively isolated, e.g., due to geographical barriers, stochastic forces reshape certain DNA characteristics of the individuals over generations such that they reflect the restricted migration and mating/reproduction patterns. Such populations are typically termed as genetically structured and they may be statistically represented in terms of several clusters between which DNA variations differ clearly from each other. When detailed knowledge of the ancestry of a natural population is lacking, the DNA characteristics of a sample of current generation individuals often provide a wealth of information in this respect. Several statistical approaches to model-based clustering of such data have been introduced, and in particular, the Bayesian approach to modeling the genetic structure of a population has attained a vivid interest among biologists. However, the possibility of utilizing spatial information from sampled individuals in the inference about genetic clusters has been incorporated into such analyses only very recently. While the standard Bayesian hierarchical modeling techniques through Markov chain Monte Carlo simulation provide flexible means for describing even subtle patterns in data, they may also result in computationally challenging procedures in practical data analysis. Here we develop a method for modeling the spatial genetic structure using a combination of analytical and stochastic methods. We achieve this by extending a novel theory of Bayesian predictive classification with the spatial information available, described here in terms of a colored Voronoi tessellation over the sample domain. Our results for real and simulated data sets illustrate well the benefits of incorporating spatial information to such an analysis.     

Filtering the histories of a partially observed marked point process

We consider a situation in which the evolution of an ‘underlying’ marked point process is of interest, but where this process is not directly observable. Instead, we assume that another marked point process, which is fully determined by the underlying process, can be observed. The problem is then the estimation, at any given time t, of the underlying development so far, given the corresponding observations. The solution, in the sense of a conditional distribution of the underlying pre-t history, is shown to satisfy a recursive filter formula. Sufficient conditions for the uniqueness of the solution are given. Two non-trivial examples are considered in detail.     

Symmetric Wiener-Hopf factorisations in Markov additive processes

The classical Wiener-Hopf factorisation of a probability measure is extended to an operator factorisation associated with a semi-Markov transition function. Some consequences of this factorisation are indicated including a set of duality relations.