Estimation and Prediction Dynamics of the Spread of the Coronavirus Infection Using Minimax Approximation Criterion and Polynomial Splines

Abstract

Presented methodology is on the spread of coronavirus infection dynamics using minimax criterion with the spline approximation. The study is based at authors mathematical method of approximation the dynamic series. The fundamental difference from classical problem of Chebyshev (1854) is the introduction of limiting conditions that allow joining splines in minimax optimality criterion mode. The approximation model is improved as a result of hierarchical procedure, at each stage of which new spline is attached. In the experiments used data on the spread of the coronavirus infection in Russia for period of more than a year with daily registration of cases of infection. So, quadratic approximation function is obtained with a negative coefficient at the highest degree of the variable, indicating a sharp decrease in the spread of coronavirus infection in Russia since mid-January 2021.