Characterizations of scale mixtures of gamma processes in terms of sufficiency and isotropy (Q1585960)

This paper aims to propose a mathematical framework for the use of scale mixtures of gamma processes as an appropriate conditional approach to limiting average rate of deterioration processes. Section 2 presents the classical characterization of gamma processes in terms of compound Poisson processes. The main results are contained in Sections 3 and 4, which provide, respectively, two characterizations of scale mixtures of gamma processes: (a) in terms of conditioning on sums of increments, serving as sufficient statistics for the unknown limiting average rate, and (b) in terms of isotropy. A useful property of a scale mixture of gamma processes is that one can always find a uniform time-partition for which the joint probability density function of the increments can be written as a scale mixture of exponentials. The scale mixtures of gamma processes are proved to be useful for modelling the deterioration processes that proceed in one direction and in random jumps, and for developing tailor-made maintenance decision models. Other papers of the authors report on the use of the investigated processes to several applications in the field of hydraulic engineering.