Short-term forecasting of daily COVID-19 cases in Brazil by using the Holt’s model

Martinez, Edson Zangiacomi; Aragon, Davi Casale; Nunes, Altacílio Aparecido

doi:10.1590/0037-8682-0283-2020

Cited by 18 publications

(16 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Martelloni and Martelloni (2020) studied the temporal evolution of the SARS-Cov-2 in Italy where among four different models; the generalized logistic model best described the situation in Italy. Martinez et al (2020) have researched 'short-term forecasting of daily COVID-19 cases in Brazil by using the Holt's model'. They have calculated MAPE (mean absolute percentage error) for each model.…”

Section: Introductionmentioning

confidence: 99%

An Analysis of COVID-19 Cases in Nepal: A Modeling Approach

Nepal

2020

J. Inst. Sci. Tech.

View full text Add to dashboard Cite

Unlike previous coronaviruses infections, COVID-19 has badly affected not only the health of people but also the socioeconomic activities of Nepal. It would help the government of Nepal to manage this crisis if a proper mechanism to predict COVID cases has been developed. This study aims to look for patterns of confirmed, recovery and death cases. Moreover, it tries to check whether Gompertz and Logistic model would be able to read the patterns of total confirmed and death cases. It also forecasts the total number of confirmed as well as death cases. Data from January 23, 2020 to October 30, 2020 obtained from the website of Wikipedia are used for analysis. Gompertz and Logistic models were fitted to the total number of confirmed and death cases and models are compared based on various criteria. Besides, an automatic ARIMA model was used to predict cumulative confirmed and death cases and the accuracy of the model was also checked. ARIMA model forecasted 347,812 confirmed cases and 1,754 death cases till December 31, 2020. At 95 % confidence interval, the confirmed cases were expected between 273,889 and 421,734 whereas death cases were estimated from 1,387 to 2,119. Both models were fitted well in both total confirmed cases and total death cases. It was found that the Logistic model fits better in total confirmed cases whereas in total death cases, the Gompertz model was better. ARIMA model precisely forecasted the number of confirmed and death cases.

show abstract

Section: Introductionmentioning

confidence: 99%

An Analysis of COVID-19 Cases in Nepal: A Modeling Approach

Nepal

2020

J. Inst. Sci. Tech.

View full text Add to dashboard Cite

show abstract

“…If they are used at the beginning of the epidemic, just to obtain a smoothened curve of the cumulative number of cases, care must be taken with the interpretation of their parameters. Short-term forecasts can be obtained from the immediate trajectories of the curves obtained from these models, which are likely to be more accurate than long-term forecasts, but are also sensitive to the high volatility observed at the end of the time series of reported cases 10 . These variations occur due to extrinsic factors, such as the availability of tests for essential screening, the natural history of the disease and changes in mitigation measures.…”

Section: /5mentioning

confidence: 99%

Long-term forecasts of the COVID-19 epidemic: a dangerous idea

Martinez

Aragon

Nunes

2020

Rev. Soc. Bras. Med. Trop.

Self Cite

View full text Add to dashboard Cite

Introduction: Mathematical models have been used to obtain long-term forecasts of the COVID-19 epidemic. Methods: The daily COVID-19 case count in two Brazilian states was used to show the potential limitations of long-term forecasting through the application of a mathematical model to the data. Results: The predicted number of cases at the end of the epidemic and at the moment that the peak occurs, is highly dependent on the length of the time series used in the predictive model. Conclusions: Predictions obtained during the course of the COVID-19 pandemic need to be viewed with caution.

show abstract

“…This late pattern may affect the prediction of long term forecast with the SIR model or with empirical logistic (Gompertz) curve models that assume a single peaked trajectory (see for example Batista, 2020;Sanchez-Villegas, 2020, Lee et al, 2020. Other more complex models based on detailed information and using either variants of nonlinear growth models (IHME, 2020) have been exposed by the specialized press to criticisms for forecast failure (underestimation) in the US (Wallace-Wells, 2020) and neural network forecast models applied to Brazil have also led to significant underestimation (Pereira et al, 2020) or have been defeated by exponential growth time series models (Martinez et al, 2020) . Finally, returning to a short term forecasting strategy one of the most complete and recognized models for forecasting COVID-19 with results for many countries (Imperial College, 2020) has also a short term horizon, centering its working on the evolution of observed deaths rather than observed cases, due to the measurement problems with the later, and using a SEIR framework.…”

Section: Introductionmentioning

confidence: 99%

COVID-19 with Uncertain Phases: Estimation Issues with An Illustration for Argentina

2020

View full text Add to dashboard Cite

We use an approach to assess COVID-19 performance that starts from what we consider is the most likely set of hypotheses about the uncertain evolution of the pandemic, that envisage a sequence of different cycles with unknown duration and magnitude over 18-24 months. This pattern implies a research strategy where shortterm time series forecasting of the evolution of observed cases and deaths play a central role in both detecting transitions from phase to phase of infections and the estimation of necessarily changing structural parameters and indicators of a SIRD model. We illustrate our approach with Buenos Aires City performance, which represents a significant share of the Argentine case with an early introduction of a lockdown followed by a second wave latter on. This approach can be extended to include measures of the intensity and compliance of lockdowns, as well as the heterogeneity across areas. We find that mobility (as a proxy for the effectiveness of the lockdown) has an impact on observed cases in Buenos Aires City with a lag of 8 days and deaths relate with new cases registered 16 to 19 days before. Mobility has a clear impact on the growth rate of cases and by extension deaths. Our approach and results have implications for policy dialogue issues.

show abstract

Short-term forecasting of daily COVID-19 cases in Brazil by using the Holt’s model

Cited by 18 publications

References 11 publications

An Analysis of COVID-19 Cases in Nepal: A Modeling Approach

An Analysis of COVID-19 Cases in Nepal: A Modeling Approach

Long-term forecasts of the COVID-19 epidemic: a dangerous idea

COVID-19 with Uncertain Phases: Estimation Issues with An Illustration for Argentina

Contact Info

Product

Resources

About