Forecasting COVID-19 cases in US states using reconstructed incidence data
Rebecca K Nash, Sangeeta Bhatia, Jack Wardle, Anne Cori, Pierre Nouvellet · 2026
Read Forecasting COVID-19 cases in US states using reconstructed incidence data by Rebecca K Nash, Sangeeta Bhatia, Jack Wardle, Anne Cori, Pierre Nouvellet (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0353983.txt.
Description
Rebecca K Nash, Sangeeta Bhatia, Jack Wardle, Anne Cori, and Pierre Nouvellet solved a practical forecasting problem: epidemic models want daily case counts, but exhausted health systems often report weekly. Their 2026 study built a statistical reconstruction method that unfolds weekly COVID-19 totals back into plausible daily series, then tested epidemic forecasts on data from thirteen American states.
Forecasts fed with reconstructed daily counts performed just as well as forecasts fed the true daily data across the 2020 to 2022 period. That parity matters for the next emergency: when daily reporting becomes unsustainable, or in places that never had it, modelers need not fly blind. Weekly reports, properly unfolded, can carry real-time prediction.
Catalog record
| CREATOR | Rebecca K Nash, Sangeeta Bhatia, Jack Wardle, Anne Cori, Pierre Nouvellet |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32073 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |