time-inhomogeneous Markov chains时间与马而可夫.pdfVIP

Lie algebra solution of population models based on time-inhomogeneous Markov chains Thomas House, Warwick Mathematics Institute 1 Abstract 1 Many natural populations are well modelled through time-inhomogeneous stochastic pro- 0 cesses. Such processes have been analysed in the physical sciences using a method based on 2 Lie algebras, but this methodology is not widely used for models with ecological, medical and v social applications. This paper presents the Lie algebraic method, and applies it to three o biologically well motivated examples. The result of this is a solution form that is often highly N computationally advantageous. 3 2 1 Introduction ] R Stochastic models based on Markov chains are important in many ecological, medical and social P contexts. In these contexts, where populations are modelled, there are often external influences h. that act on the system in a manner that varies over time, leading to a time-inhomogeneous Markov t chain model and corresponding technical difficulties for analysis [10]. a Wei and Norman [14] proposed a method for dealing analytically with time-inhomogeneous m Markov chains, based on Lie algebraic methods. The idea of combining Lie algebras and symmetry [ considerations with Markov chains has continued to attract theoretical interest in a variety of 1 contexts [15, 6, 9, 12]. v At the same time, there is a more applied desire to have numerically efficient methods to analyse