Parameter Estimation of Multi-Staged SI(n)RS Epidemic Models

Abstract

This paper designs an algorithm to estimate the parameters of a discrete-time SI(n)RS multi-staged model. Initially, the structural properties of the discrete-time model, such as well-posedness and positivity, are studied. The estimation procedure is based on multivariable least-squares and contains a projection algorithm in order to ensure that the estimated parameters are always nonnegative, in accordance to their physical meaning. A simulation example shows the usefulness of the presented approach.