The conversion rate between asymptomatic infections and reported/unreported symptomatic infections is a very sensitive parameter for model variables that spread COVID-19. This evaluation shows that the proposed approach achieves a good predictive power for out-of-sample predictions within one week (R-squared > 0.9 at 1 day, R-squared > 0.7 at 7 days), while it decreases with increasing forecasted days (R-squared > 0.5 at 14 days). Starting from the database of the Italian Civil Protection Department (DPC), the experimental evaluation was performed on real-world data collected from February 2020 to March 2022, focusing on Calabria, a region of Southern Italy. This paper presents the design and implementation of an approach based on autoregressive models to reliably forecast the spread of COVID-19 in Italian regions. For the COVID-19, accurate and timely forecasts are therefore essential to provide the opportunity to rapidly identify risk areas affected by the pandemic, reallocate the use of health resources, design countermeasures and increase public awareness. Despite the stunning speed with which highly effective and safe vaccines have been developed, the emergence of new variants of SARS‑CoV‑2 causes high rates of (re)infection, a major impact on health care services, and a slowdown to the socio-economic system.
0 kommentar(er)
