Published online by Cambridge University Press: 14 July 2016
In this article we consider a generalisation of conservative processes in which the usual transition rate parameters are replaced by time-dependent stochastic variables. The main result of the article shows that these generalised processes which we call conservative processes with stochastic rates have transition probabilities which can be characterised in terms of exchangeable random variables in a manner similar to the characterisation of conservative processes in terms of independent random variables given by Bartlett (1949). We use this characterisation to obtain general expressions for the transition probabilities and to examine some limiting aspects of the processes. The carrier-borne epidemic is treated as a particular case of these generalised processes.
This paper was presented at the Conference on Biomathematics and Biostatistics, Department of Pure and Applied Mathematics, Washington State University, Pullman, Washington, 1 and 2 May, 1974.