Assessment of COVID-19 hospitalization forecasts from a simplified SIR model

TL;DR

This paper introduces the SH model, a simplified two-parameter SIR model, capable of accurately forecasting COVID-19 hospitalizations over months with minimal parameters, demonstrating high predictive accuracy in Belgium and some French departments.

Contribution

The paper presents a simplified, two-parameter SIR model that effectively forecasts COVID-19 hospitalizations with high accuracy, reducing complexity compared to traditional models.

Findings

High accuracy in fitting COVID-19 hospitalization data for Belgium.

Forecasts with less than 4% MAPE for two months after training.

Successful predictions in 14 French departments with MAPE below 20%.

Abstract

We propose the SH model, a simplified version of the well-known SIR compartmental model of infectious diseases. With optimized parameters and initial conditions, this time-invariant two-parameter two-dimensional model is able to fit COVID-19 hospitalization data over several months with high accuracy (e.g., the root relative squared error is below 10% for Belgium over the period from 2020-03-15 to 2020-07-15). Moreover, we observed that, when the model is trained on a suitable three-week period around the first hospitalization peak for Belgium, it forecasts the subsequent two months with mean absolute percentage error (MAPE) under 4%. We repeated the experiment for each French department and found 14 of them where the MAPE was below 20%. However, when the model is trained in the increase phase, it is less successful at forecasting the subsequent evolution.