In 2020, America experienced a devastating outbreak of COVID-19. Tracking the spread and epidemic curves of this virus at the county level is critical for effective local responses to outbreaks. We attempt to solve this problem by combining county-level socioeconomic, epidemiological, and mobility datasets with state-level case projections and testing data to train machine learning models to predict epidemic curves. Through our collaborations with the COVID-19 Forecast Hub and the NIGMS MIDAS Network, our data is shared with the US CDC to inform our national response.
We are grateful to all the wonderful institutions and individuals whose hard work produces the datasets we use. For a list of these, please check out our home page.