An Assessment of the IHME Covid-19 Model: US Fatalities in 2020

Authors

  • Thomas R. Robbins East Carolina University

DOI:

https://doi.org/10.33423/ajm.v21i2.4225

Keywords:

management, COVID-19, forecasting, forecast accuracy, health metrics

Abstract

The Covid-19 forecasting model published by the Institute for Health Metrics and Evaluation (IHME) is one of the most influential, consequential, and controversial forecasting models ever published. The model has been widely cited by policy makers, but it has also been widely criticized. In this paper we make an objective, external review of the model. We evaluate the outputs of the model, how they changed, and how they compared to actual results. The model was subject to frequent adjustment that resulted in wide swings back and forth. The errors associated with the model were high and actual results fell inside the model’s reported 95% confidence interval far less than 95% of the time. Overall, we find the accuracy of the model was poor, and the model’s predictions were unrealistically precise.

Downloads

Published

2021-06-30

How to Cite

Robbins , T. R. . (2021). An Assessment of the IHME Covid-19 Model: US Fatalities in 2020 . American Journal of Management, 21(2). https://doi.org/10.33423/ajm.v21i2.4225

Issue

Section

Articles