Stochastic SIR model including growths: modeling and inference
In this talk the focus is on a Susceptible-Infected-Removed stochastic model in which the stochasticity is introduced by means of two independent Brownian motions in the dynamics of Susceptible and Infected populations. A growth function is also considered in the Susceptible population in order to model the natural evolution of the population in which the size is influenced by the births and deaths. The inference for such a model is addressed by means of a quasi-maximum likelihood method. The resulting nonlinear system is numerically solved by an iterative procedure. A technique to obtain the initial solutions usually required by numerical techniques is also provided. Simulation studies for several growth functions in the susceptible population are discussed in order to show the performances of the initial estimates of the involved parameters, as well as to evaluate the goodness of the proposed methodology.
Keywords: Euler-Maruyama scheme growth curves inference Newton method quasi-maximum likelihood estimation