Testing the predictive accuracy of COVID-19 forecasts

Laura Coroneo; Fabrizio Iacone; Alessia Paccagnini; Paulo Santos Monteiro

doi:10.1016/j.ijforecast.2022.01.005

Testing the predictive accuracy of COVID-19 forecasts

Laura Coroneo, Fabrizio Iacone, Alessia Paccagnini, Paulo Santos Monteiro

Economics

Research output: Contribution to journal › Article › peer-review

Abstract

We test the predictive accuracy of forecasts of the number of COVID-19 fatalities produced by several forecasting teams and collected by the United States Centers for Disease Control and Prevention for the epidemic in the United States. We find three main results. First, at the short horizon (1-week ahead) no forecasting team out performs a simple time-series benchmark. Second, at longer horizons (3- and 4-week ahead)forecasters are more successful and sometimes outperform the benchmark. Third, one of the best performing forecasts is the Ensemble forecast, that combines all available predictions using uniform weights. In view of these results, collecting a wide range of forecasts and combining them in an ensemble forecast may be a superior approach for health authorities, rather than relying on a small number of forecasts.

Original language	English
Pages (from-to)	606-622
Number of pages	17
Journal	International journal of forecasting
Volume	39
Early online date	4 Mar 2023
DOIs	https://doi.org/10.1016/j.ijforecast.2022.01.005
Publication status	E-pub ahead of print - 4 Mar 2023

Bibliographical note

© 2022 Published by Elsevier B.V. on behalf of International Institute of Forecasters. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.

Keywords

Forecast evaluation, Forecasting tests, Epidemic

Access to Document

10.1016/j.ijforecast.2022.01.005Licence: CC BY-NC-ND

Embargoed Document

Covid_IJoF2
Accepted author manuscript, 586 KB
Licence: CC BY-NC-ND
Embargo ends: 4/03/25

Cite this

@article{329b32781e7b47d386d30287276634ef,

title = "Testing the predictive accuracy of COVID-19 forecasts",

abstract = "We test the predictive accuracy of forecasts of the number of COVID-19 fatalities produced by several forecasting teams and collected by the United States Centers for Disease Control and Prevention for the epidemic in the United States. We find three main results. First, at the short horizon (1-week ahead) no forecasting team out performs a simple time-series benchmark. Second, at longer horizons (3- and 4-week ahead)forecasters are more successful and sometimes outperform the benchmark. Third, one of the best performing forecasts is the Ensemble forecast, that combines all available predictions using uniform weights. In view of these results, collecting a wide range of forecasts and combining them in an ensemble forecast may be a superior approach for health authorities, rather than relying on a small number of forecasts.",

keywords = "Forecast evaluation, Forecasting tests, Epidemic",

author = "Laura Coroneo and Fabrizio Iacone and Alessia Paccagnini and {Santos Monteiro}, Paulo",

note = "{\textcopyright} 2022 Published by Elsevier B.V. on behalf of International Institute of Forecasters. This is an author-produced version of the published paper. Uploaded in accordance with the publisher{\textquoteright}s self-archiving policy. ",

year = "2023",

month = mar,

day = "4",

doi = "10.1016/j.ijforecast.2022.01.005",

language = "English",

volume = "39",

pages = "606--622",

journal = "International journal of forecasting",

issn = "0169-2070",

publisher = "Elsevier",

}

TY - JOUR

T1 - Testing the predictive accuracy of COVID-19 forecasts

AU - Coroneo, Laura

AU - Iacone, Fabrizio

AU - Paccagnini, Alessia

AU - Santos Monteiro, Paulo

N1 - © 2022 Published by Elsevier B.V. on behalf of International Institute of Forecasters. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy.

PY - 2023/3/4

Y1 - 2023/3/4

N2 - We test the predictive accuracy of forecasts of the number of COVID-19 fatalities produced by several forecasting teams and collected by the United States Centers for Disease Control and Prevention for the epidemic in the United States. We find three main results. First, at the short horizon (1-week ahead) no forecasting team out performs a simple time-series benchmark. Second, at longer horizons (3- and 4-week ahead)forecasters are more successful and sometimes outperform the benchmark. Third, one of the best performing forecasts is the Ensemble forecast, that combines all available predictions using uniform weights. In view of these results, collecting a wide range of forecasts and combining them in an ensemble forecast may be a superior approach for health authorities, rather than relying on a small number of forecasts.

AB - We test the predictive accuracy of forecasts of the number of COVID-19 fatalities produced by several forecasting teams and collected by the United States Centers for Disease Control and Prevention for the epidemic in the United States. We find three main results. First, at the short horizon (1-week ahead) no forecasting team out performs a simple time-series benchmark. Second, at longer horizons (3- and 4-week ahead)forecasters are more successful and sometimes outperform the benchmark. Third, one of the best performing forecasts is the Ensemble forecast, that combines all available predictions using uniform weights. In view of these results, collecting a wide range of forecasts and combining them in an ensemble forecast may be a superior approach for health authorities, rather than relying on a small number of forecasts.

KW - Forecast evaluation, Forecasting tests, Epidemic

U2 - 10.1016/j.ijforecast.2022.01.005

DO - 10.1016/j.ijforecast.2022.01.005

M3 - Article

SN - 0169-2070

VL - 39

SP - 606

EP - 622

JO - International journal of forecasting

JF - International journal of forecasting

ER -