A fractional model for the COVID-19 pandemic: Application to Italian data

TL;DR

This paper introduces a probabilistic SIRD model with fractional Brownian motion to better capture the randomness in COVID-19 infection, recovery, and death rates in Italy, requiring minimal parameters for calibration.

Contribution

It presents a novel fractional stochastic SIRD model that incorporates randomness in key rates, enhancing the modeling of COVID-19 dynamics with fewer parameters.

Findings

Model effectively captures COVID-19 spread in Italy.

Incorporating fractional Brownian motion improves fit to data.

Model's simplicity allows easy calibration.

Abstract

We provide a probabilistic SIRD model for the COVID-19 pandemic in Italy, where we allow the infection, recovery and death rates to be random. In particular, the underlying random factor is driven by a fractional Brownian motion. Our model is simple and needs only some few parameters to be calibrated.