Curating a COVID-19 Data Repository and Forecasting County-Level Death Counts in the United States

• Activity Number: 300 - Addressing Statistical Challenges of the COVID-19 Crisis
• Type: Invited
• Date/Time: Wednesday, August 11, 2021 : 3:30 PM to 5:20 PM
• Sponsor: ENAR
• Author(s): Nick Altieri and Rebecca L Barter and James Duncan and Raaz Dwivedi and Karl Kumbier and Xiao Li and Robert Netzorg and Briton Park and Chandan Singh and Yan Shuo Tan and Tiffany Tang* and Yu Wang and Chao Zhang and Bin Yu
• Companies: University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, San Francisco and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Berkeley and University of California, Santa Barbara and University of California, Berkeley
• Keywords: COVID-19; data repository; time series forecasting; ensemble methods; prediction intervals

Abstract

As the COVID-19 outbreak evolves, accurate forecasting continues to play an important role in informing policy decisions. In this work, we present our curation of a large COVID-19 data repository with 20+ data sources and counting. Using this data, we develop an ensemble prediction method called Combined Linear and Exponential Predictors (CLEP) for forecasting the short-term trajectory of COVID-19 cumulative death counts at the county-level in the US up to two weeks ahead. We also use the prediction errors from the past five days to assess the uncertainty of CLEP, resulting in generally-applicable prediction intervals, Maximum (absolute) Error Prediction Intervals (MEPI). MEPI achieves a coverage rate of more than 94% when averaged across counties for predicting cumulative death counts two weeks in the future. Our forecasts are currently being used by the non-profit organization, Response4Life, and have directly contributed to the distribution of medical supplies across the country. We hope that our forecasts and data repository at https://covidseverity.com can help guide necessary county-specific decision-making and help counties prepare for their continued fight against COVID-19.