A stochastic model of the HIV epidemic and the HIV infection distribution in a homosexual population.

In this paper we develop a stochastic model for the HIV epidemic in a homosexual population and use the model to characterize the HIV infection distribution and seroconversion distribution. Through computer-generated infection distributions and seroconversion distributions, we assess the effects of various risk factors on these distributions. The fitting of some data sets generated by computer suggests that the three-parameter generalized log-logistic distribution should be assumed as the infection distribution for the proposed stochastic model of HIV epidemics.