Dataset Information

Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil.

ABSTRACT: The new Coronavirus (COVID-19) is an emerging disease responsible for infecting millions of people since the first notification until nowadays. Developing efficient short-term forecasting models allow forecasting the number of future cases. In this context, it is possible to develop strategic planning in the public health system to avoid deaths. In this paper, autoregressive integrated moving average (ARIMA), cubist regression (CUBIST), random forest (RF), ridge regression (RIDGE), support vector regression (SVR), and stacking-ensemble learning are evaluated in the task of time series forecasting with one, three, and six-days ahead the COVID-19 cumulative confirmed cases in ten Brazilian states with a high daily incidence. In the stacking-ensemble learning approach, the CUBIST regression, RF, RIDGE, and SVR models are adopted as base-learners and Gaussian process (GP) as meta-learner. The models' effectiveness is evaluated based on the improvement index, mean absolute error, and symmetric mean absolute percentage error criteria. In most of the cases, the SVR and stacking-ensemble learning reach a better performance regarding adopted criteria than compared models. In general, the developed models can generate accurate forecasting, achieving errors in a range of 0.87%-3.51%, 1.02%-5.63%, and 0.95%-6.90% in one, three, and six-days-ahead, respectively. The ranking of models, from the best to the worst regarding accuracy, in all scenarios is SVR, stacking-ensemble learning, ARIMA, CUBIST, RIDGE, and RF models. The use of evaluated models is recommended to forecasting and monitor the ongoing growth of COVID-19 cases, once these models can assist the managers in the decision-making support systems.

SUBMITTER: Ribeiro MHDM

PROVIDER: S-EPMC7252162 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil.

Ribeiro Matheus Henrique Dal Molin MHDM da Silva Ramon Gomes RG Mariani Viviana Cocco VC Coelho Leandro Dos Santos LDS

Chaos, solitons, and fractals 20200501

The new Coronavirus (COVID-19) is an emerging disease responsible for infecting millions of people since the first notification until nowadays. Developing efficient short-term forecasting models allow forecasting the number of future cases. In this context, it is possible to develop strategic planning in the public health system to avoid deaths. In this paper, autoregressive integrated moving average (ARIMA), cubist regression (CUBIST), random forest (RF), ridge regression (RIDGE), support vecto ...[more]

PMID: 32501370

Similar Datasets

Project description:The outbreak of coronavirus disease 2019 (COVID-19), caused by the virus SARS-CoV-2, has been rapidly increasing in the United States. Boroughs of New York City, including Queens county, turn out to be the epicenters of this infection. According to the data provided by the New York State Department of Health, most of the cases of new COVID-19 infections in New York City have been found in the Queens county where 42,023 people have tested positive, and 3221 people have died as of 20 April 2020. Person-to-person transmission and travels were implicated in the initial spread of the outbreaks, but factors related to the late phase of rapidly spreading outbreaks in March and April are still uncertain. A few previous studies have explored the links between air pollution and COVID-19 infections, but more data is needed to understand the effects of short-term exposures of air pollutants and meteorological factors on the spread of COVID-19 infections, particularly in the U.S. disease epicenters. In this study, we have focused on ozone and PM2.5, two major air pollutants in New York City, which were previously found to be associated with respiratory viral infections. The aim of our regression modeling was to explore the associations among ozone, PM2.5, daily meteorological variables (wind speed, temperature, relative humidity, absolute humidity, cloud percentages, and precipitation levels), and COVID-19 confirmed new cases and new deaths in Queens county, New York during March and April 2020. The results from these analyses showed that daily average temperature, daily maximum eight-hour ozone concentration, average relative humidity, and cloud percentages were significantly and positively associated with new confirmed cases related to COVID-19; none of these variables showed significant associations with new deaths related to COVID-19. The findings indicate that short-term exposures to ozone and other meteorological factors can influence COVID-19 transmission and initiation of the disease, but disease aggravation and mortality depend on other factors.

Project description:AbstractThe coronavirus (COVID-19) disease outbreak was a public health emergency of international concern which eventually evolved into a pandemic. Nigeria was locked down in March, 2020 as the country battled to contain the spread of the disease. By August 2020, phase-by-phase easing of the lockdown was commenced and university students will soon return for academic activities. This study undertakes some epidemiological analysis of the Nigerian COVID-19 data to help the government and university administrators make informed decisions on the safety of personnel and students.The COVID-19 data on confirmed cases, deaths, and recovered were obtained from the website of the Nigerian Centre for Disease Control (NCDC) from April 2, 2020 to August 24, 2020. The infection rate, prevalence, ratio, cause-specific death rate, and case recovery rate were used to evaluate the epidemiological characteristics of the pandemic in Nigeria. Exponential smoothing was adopted in modeling the time series data and forecasting the pandemic in Nigeria up to January 31, 2021.The results indicated that the pandemic had infection rate of at most 3 infections per 1 million per day from April to August 2020. The death rate was 5 persons per 1 million during the period of study while recovery rate was 747 persons per 1000 infections. Analysis of forecast data showed steady but gradual decrease in the daily infection rate and death rate and substantial increase in the recovery rate, 975 recoveries per 1000 infections.In general, the epidemiological attributes of the pandemic from the original data and the forecast data indicated optimism in the decrease in the rate of infection and death in the future. Moreover, the infection rate, prevalence and death rate in January 2021 coincided with the predictions based on the analysis. Therefore, the Nigerian government is encouraged to allow universities in the country to reopen while university administrators set up the necessary protocols for strict adherence to safety measures.

Dataset Information

Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil.

Publications

Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets