[Paper Review] Propagation analysis and prediction of the COVID-19

Lixiang Li, Zihang Yang+8 more|arXiv (Cornell University)|Mar 15, 2020

COVID-19 epidemiological studies9 references204 citations

TL;DR

The paper models COVID-19 transmission using official data, achieving an error within 3% between the model and observed curves, and demonstrates forward prediction and backward inference of epidemic dynamics.

ABSTRACT

Based on the official data modeling, this paper studies the transmission process of the Corona Virus Disease 2019 (COVID-19). The error between the model and the official data curve is within 3%. At the same time, it realized forward prediction and backward inference of the epidemic situation, and the relevant analysis help relevant countries to make decisions.

Motivation & Objective

• Motivate timely understanding of COVID-19 transmission dynamics using official data.
• Develop a data-driven transmission model with accurate fit to observed case curves.
• Enable forward forecasting and backward inference to support policy decisions.

Proposed method

• Use official data modeling to study the transmission process of COVID-19.
• Quantify the error between the model curve and official data as within 3%.
• Perform forward prediction and backward inference of the epidemic dynamics.

Experimental results

Research questions

• RQ1 How accurately can official data be used to model the transmission dynamics of COVID-19?
• RQ2 Can the model provide reliable forward predictions of epidemic trajectories?
• RQ3 Is backward inference feasible to reconstruct past epidemic states from observed data?

Key findings

• The error between the model and official data curve is within 3%.
• The approach enables forward prediction of the epidemic trajectory.
• The approach enables backward inference of the epidemic situation.