Abstract Hannah Biegel

Abstract: Forecasting seasonal influenza using data assimilation methods

At the onset of an epidemic, parameters for mathematical models are often not known and the available data are frequently noisy. Here we describe a simple model for seasonal influenza and use variational data assimilation (VDA) to systematically address the issue of imperfect data and to fit model parameters. We first test the efficacy of VDA on the model near the onset of the influenza season using so-called "synthetic data experiments." We then apply these techniques to CDC data from recent influenza seasons.