On evaluation of infection probabilities for different age groups

摘要

It is well known that evaluation of parameters for any mathematical model is always very important and often a very difficult problem. This is especially true for agents-based models because as a rule the evaluation of an agent's parameters can be made only based on available information from the higher levels of a complex system. Such problems can be ill-posed or even have a non-unique solution. That is why there is no general algorithm for solving these problems and a researcher has to search for it for every specific model type. In this paper the problem of evaluating transmission probabilities for different age groups during an influenza epidemic/pandemic in a city will be considered. In order to solve this problem a regularization procedure based on illness attack rates for several age groups (cumulated numbers of infected residents as functions of time) is proposed and validated. It can be used for past epidemics to estimate the efficiency or inefficiency of undertaken interventions, to propose new ones and to reveal its advantages and shortcomings. Under some conditions it can be used to model the possible dynamics of a coming epidemic.