Back

Forecasting COVID-19 new cases in Algeria using Autoregressive fractionally integrated moving average Models (ARFIMA)

Balah, B.; Djeddou, M.

2020-05-08 epidemiology
10.1101/2020.05.03.20089615 medRxiv
Show abstract

In this research, an ARFIMA model is proposed to forecast new COVID-19 cases in Algeria two weeks ahead. In the present study, public health database from Algeria health ministry has been used to build an ARFIMA model and used to forecast COVID-19 new cases in Algeria until May 11, 2020. BackgroundThe aim of this study is first to find the best prediction method among the two techniques used and type of memory, either short or long, of the model constructed for the daily confirmed cases in Algeria, then make forecasts of the confirmed cases in the fifteen next days. MethodsThis study was conducted based on daily new cases of COVID-19 that were collected from the official website of Algerian Ministry of Health from March 1, 2020 to April 26, 2020. Auto Regressive Integrated Moving Average (ARFIMA) model was used to predict the trend of confirmed cases. The evaluation of the fractional differentiation parameter (d) is carried out using OxMetrics 6 software. ResultsThe ARFIMA model (0, 0.431779, 0) build for Algeria, has a long memory and an upward trend over the next fifteen days and which coincides with the holy month of Ramadhan. ConclusionsThe forecasted results obtained by the proposed ARFIMA model can be used as a decision support tool to manage medical efforts and facilities against the COVID-19 pandemic crisis.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.