A non-central beta model to forecast and evaluate pandemics time series

Abstract

Government, researchers, and health professionals have been challenged to model, forecast, and evaluate pandemics time series (e.g. new coronavirus SARS-CoV-2, COVID-19). The main difficulty is the level of novelty imposed by these phenomena. Information from previous epidemics is only partially relevant. Further, the spread is local-dependent, reflecting a number of social, political, economic, and environmental dynamic factors. The present paper aims to provide a relatively simple way to model, forecast, and evaluate the time incidence of a pandemic. The proposed framework makes use of the non-central beta (NCB) probability density function. Specifically, a probabilistic optimisation algorithm searches for the best NCB model of the pandemic, according to the mean square error metric. The resulting model allows one to infer, among others, the general peak date, the ending date, and the total number of cases as well as to compare the level of difficult imposed by the pandemic among territories. Case studies involving COVID-19 incidence time series from countries around the world suggest the usefulness of the proposed framework in comparison with some of the main epidemic models from the literature (e.g. SIR, SIS, SEIR) and established time series formalisms (e.g. exponential smoothing - ETS, autoregressive integrated moving average - ARIMA).

Dynamics diagnosis of the COVID-19 deaths using the Pearson diagram

“…Furthermore, we clarify that cluster number (three) in our empirical evidence is associated with the four phases of COVID-19. For more details, see: [23] , [24] , [25] , [26] . In this way, the discussion is divided into two distinct phases.…”

Optimization in the Context of COVID-19 Prediction and Control: A Literature Review

“…Finally, some studies [16] – [22] , [131] – [146] followed a model-agnostic approach and relied solely on the historical time series data of COVID-19 cases or other relevant predictors to forecast future cases. These methods employ machine learning models (neural networks [17] – [21] , [133] , [135] , [138] , [142] and deep learning [139] ) to make predictions while using various optimization algorithms (such as Gaussian process regression [16] , Bayesian optimization [17] , and metaheuristic algorithms [18] – [22] , [144] , [147] – [149] ) to optimize the model hyperparameters.…”

Predictability of COVID-19 worldwide lethality using permutation-information theory quantifiers

“…Countries and territories suffered different impacts due to the SARS-COV-2 pandemic reflected by the cumulative deaths per 100 thousand population that varied enormously from 0.0 in Mongolia to 191.5 in San Marino [7] . Previous studies indicate different spread dynamics between diverse countries, maybe reflecting different degrees of efficiency in relation to the response to the pandemic [8] , [9] . It is essential to understand COVID-19 lethality dynamics to design more robust strategies to the pandemic.…”