Analysis of the COVID-19 pandemic by SIR model and machine learning technics for forecasting

Ndiaye, Babacar Mbaye; Tendeng, Lena; Seck, Diaraf

doi:10.48550/arxiv.2004.01574

Cited by 13 publications

(19 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where τ denotes transmissibility and c contact rate and where the infectious period i equals one over the recovery rate γ. This relationship holds for well-mixed populations, as assumed by standard compartmental models like SIR or SEIR which apply the law of mass action [4,5,6,7]. They are also valid for small-to-medium spatial scales.…”

Section: Population Densitymentioning

confidence: 84%

See 1 more Smart Citation

Impact of contamination factors on the COVID-19 evolution in Senegal

Ndiaye,

Sarr,

Ndiaye

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this article, we perform an analysis of COVID-19 on one of the South Saharan countries (hot zone), the Senegal (West Africa). Many questions remain unanswered: why the African continent is not very contaminated compared to other continents. Factors of cross immunity, temperature, population density, youth, etc. are taken into account for an analysis of the contamination factors. Numerical simulations are carried out for a prediction over the coming week.

show abstract

Section: Population Densitymentioning

confidence: 84%

“…An interdisciplinary approach seeks to validate new models inspired by the African context. Based to our analysis, we can propose in future work, the generalized SIR model [4,5,6,7] taking into account to all these factors (cross immunity, youth, etc.) to analysis the coronavirus pandemic (COVID-19).…”

Section: Discussionmentioning

confidence: 99%

Impact of contamination factors on the COVID-19 evolution in Senegal

Ndiaye,

Sarr,

Ndiaye

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The objective of this work is to model the variables potentially involved in the spread of cases resulting from community transmission of COVID-19 in Senegal in order to identify statistical associations. In our previous articles [4,5,6,7,8], we try to forecast the pandemic for Senegal using the SIR, stochastic SIR and machine learning. Here, we give forecasting pandemic size of community cases for Senegal and daily predictions using the logistic model.…”

Section: Introductionmentioning

confidence: 99%

Logistic growth model and modeling of factors for community case transmission

Diouf¹,

Ndiaye²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

In this article, we analyze the spread of cases resulting from community transmission of COVID-19 in Senegal in order to identify statistical associations. The identification and knowledge of the factors associated with this community transmission can be a decision support tool to limit the spread of the disease. We estimate parameters and evaluate the growth factor, community rate, weekly increase and daily difference, and make forecasting to help on how to find concrete actions to control the situation.

show abstract

“…Finally, many approaches in the realm of machine learning have been developed [5], including e.g. Facebook's prophet algorithm [40], gradient boosted trees [47], or neural networks [55]. This list only covers a small fraction of published models, an exemplary overview of others is also given in [3,20,33].…”

Section: Introductionmentioning

confidence: 99%

Backtesting the predictability of COVID-19

Gordeev¹,

Singer²,

Michailidis³

et al. 2020

Preprint

View full text Add to dashboard Cite

The advent of the COVID-19 pandemic has instigated unprecedented changes in many countries around the globe, putting a significant burden on the health sectors, affecting the macro economic conditions, and altering social interactions amongst the population through a number of mitigation measures and governmental instructions. In response, the academic community has produced multiple forecasting models, approaches and algorithms to best predict the different indicators of COVID-19, such as the number of confirmed infected cases, the number of deceased and economic indicators. Specifically at the beginning of the pandemic, researchers had little to no historical information about the pandemic at their disposal in order to inform their forecasting methods. Our work studies the predictive performance of models at various stages of the pandemic to better understand their fundamental uncertainty and the impact of data availability on such forecasts. We use historical data of infected, deceased and recovered cases of COVID-19 from 253 regions from the period of 22nd January 2020 until 22nd June 2020 to predict, through a rolling window backtesting framework, the cumulative number of infected cases for the next 7 and 28 days. We implement three simple models to track the root mean squared logarithmic error in this 6-month span, a baseline model that always predicts the last known value of the cumulative confirmed cases, a power growth model and an epidemiological model called SEIRD. Within our presented backtesting framework, each model is re-fitted daily and its best parameters are obtained with information known as of that point in time (in terms of historical confirmed, deceased and recovered cases) and predictions are extended for the 7-days and 28-days forecast horizons. We demonstrate that prediction errors are substantially higher in early stages of the pandemic, resulting from limited data. Throughout the course of the pandemic, errors regress slowly, but steadily. The more confirmed cases a country exhibits at any point in time, the lower the error in forecasting future confirmed cases. Our work emphasizes the significance of having a rigorous backtesting framework to accurately assess the predictive power of such models at any point in time during the outbreak which in turn can be used to assign the right level of certainty to these forecasts and facilitate better planning.

show abstract

Analysis of the COVID-19 pandemic by SIR model and machine learning technics for forecasting

Cited by 13 publications

References 0 publications

Impact of contamination factors on the COVID-19 evolution in Senegal

Impact of contamination factors on the COVID-19 evolution in Senegal

Logistic growth model and modeling of factors for community case transmission

Backtesting the predictability of COVID-19

Contact Info

Product

Resources

About