Forecasting COVID-19 Disease Cases Using the SARIMA-NNAR Hybrid Model

  1. İbrahim DEMİR
  2. Murat KİRİSCİ

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background: COVID-19 is a new disease that is associated with high morbidity that has spread around the world. Credible estimating is crucial for control and prevention. Nowadays, hybrid models have become popular, and these models have been widely implemented. Better estimation accuracy may be attained using time-series models. Thus, our aim is to forecast the number of COVID-19 cases with time-series models. Objective: Using time-series models to predict deaths due to COVID-19. Design: SARIMA, NNAR, and SARIMA-NNAR hybrid time series models were used using the COVID-19 information of the Republic of Turkey Health Ministry. Participants: We analyzed data on COVID-19 in Turkey from March 11, 2020 to February 22, 2021. Main Measures: Daily numbers of COVID-19 confirmed cases and deaths. Materials and methods: We fitted a seasonal autoregressive integrated moving average (SARIMA)–neural network nonlinear autoregressive (NNAR) hybrid model with COVID-19 monthly cases from March 11, 2020, to February 22, 2021, in Turkey. Additionally, a SARIMA model, an NNAR model, and a SARIMA–NNAR hybrid model were established for comparison and estimation. Results The RMSE, MAE, and MAPE values of the NNAR model were obtained the lowest in the training set and the validation set. Thus, the NNAR model demonstrates excellent performance whether in fitting or forecasting compared with other models. Conclusions The NNAR model that fits this study is the most suitable for estimating the number of deaths due to COVID-19. Hence, it will facilitate the prevention and control of COVID-19.

Article activity feed

  1. Version published to 10.32323/ujma.1010490
  2. ScreenIT

    SciScore for 10.1101/2021.04.26.21256108: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Ethicsnot detected.
    Sex as a biological variablenot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    However, there are some limitations to this study. The first of these limitations is where the original data is taken. These data were obtained from the COVID-19 Information Page of the Republic of Turkey Health Ministry. This data can be included the possibility of false reporting and negligent reporting. The quality of the data can affect the build process and performance of the model to some extent. However, in order to obtain a better model, the best model was tried to be obtained by examining the p and q values from 1 to 10 instead of using an automatic model in the NNAR model. This model could also be achieved with SARIMA. In general, this process has not been tried, as the NNAR model gives better results than the SARIMA model.

    Results from TrialIdentifier: We found the following clinical trial numbers in your paper:

    IdentifierStatusTitle
    NCT04678193Enrolling by invitationCOVID-19 Risk Assessment for Hospitalization Outcomes and Ep…

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    Results from scite Reference Check: We found no unreliable references.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  3. Version published to 10.1101/2021.04.26.21256108v1 on medRxiv