The University of Texas COVID-19 Modeling Consortium
COVID-19 Mortality Projections for US States
These graphs show both the reported and projected number of COVID-19 deaths per day across the US and for individual states. We use local data from mobile-phone GPS traces to quantify the changing impact of social-distancing measures on “flattening the curve.”
Code syntax and daily updates of our forecasts are available on our UT-COVID GitHub repository.
For more information, please visit our model FAQ.
Note: On May 7, 2020 we switched to using data on confirmed and probable COVID-19 deaths to make forecasts, rather than only confirmed deaths.
Key model assumptions: The model estimates the extent of social distancing using geolocation data from mobile phones and assumes that the extent of social distancing does not change during the period of forecasting. On June 2, 2020 we started reporting an ensemble forecast, combining predictions from our original model, based on "curve-fitting," and a new SEIR-based model. On June 11, 2020, we started reporting predictions based solely on the SEIR-based model. We are currently working on a technical report to accompany this model.
Recently, JHU CSSE updated their COVID-19 mortality data for Texas to include previously uncounted deaths. A large number of additional deaths was added on July 27, 2020, although these deaths actually occurred over several weeks. To avoid bias in our projections, we manually changed the data point for July 27 to the average number of deaths across the preceding seven days.