an:05158556
Zbl 1114.62027
Williams, Brian J.; Huzurbazar, Aparns V.
Posterior sampling with constructed likelihood functions: an application to flowgraph models
EN
Appl. Stoch. Models Bus. Ind. 22, No. 2, 127-137 (2006).
00125881
2006
j
62F15 65C40 62P30
Bayesian predictive distribution; censored data; flowgraph model; incomplete data; slice sampling; queuing model
The authors consider posterior sampling in situations where data are incomplete in such a way that likelihood functions corresponding to portions of the data must be constructed. Such situations arise in the modelling of time-to-event data when not all of the event occurrences are observed. The estimation of Bayesian predictive distributions for flowgraph models using Laplace transform inversion and slice sampling techniques are described. The authors construct likelihood functions for the incomplete data and use them in a Markov chain Monte Carlo algorithm to sample from the approximate posterior and compute Bayes predictive distributions. A real data example for a cellular telephone network is used.
Aleksandr D. Borisenko (Ky??v)