Projects

    There are 10 public projects and 3 projects with access limited to collaborators only.

    Project Owner Description Summary
    CDC FluSight Forecast Hub
    lshandross Forecasts of confirmed influenza hospitalization admissions during the 2023–2024 influenza season, created as part of the US CDC’s collaborative influenza forecasting exercise
    43 models, 864 forecasts, 276,366 predictions
    COVID-19 Forecasts
    covid19hub The goal of this repository is to create a standardized set of data on forecasts from experienced teams making projections of cumulative and incident deaths due to COVID-19 in the United States.
    129 models, 9106 forecasts, 113.4 million predictions
    CDC Influenza Hospitalization Forecasts
    lshandross Forecasts of confirmed influenza hospitalization admissions during the 2021-2022 and 2022–2023 influenza seasons, created as part of the CDC’s influenza forecasting exercise "FluSight".
    44 models, 1405 forecasts, 454,334 predictions
    COVID-19 Forecasts Viz Test
    cornell The goal of this repository is to create a standardized set of data on forecasts from experienced teams making projections of cumulative and incident deaths due to COVID-19 in the United States.
    Two models, 10 forecasts, 80 predictions
    ECDC European COVID-19 Forecast Hub (No logo) epiforecasts The goal of this repository is to create a standardized set of data on forecasts from experienced teams making projections of incident case and death numbers due to COVID-19 in the EU and the United Kingdom.
    69 models, 2514 forecasts, 496,059 predictions
    Aggregating Statistical Models and Human Judgment
    allicodi This project aims to provide public health officials forecasts of the COVID-19 outbreak using human judgment.

    A consensus forecast is an equally weighted combination of probabilistic predictions made by individuals. We included consensus forecasts from individual forecasters on Metaculus (https://www.metaculus.com/questions/) and from Good Judgment Open (https://www.gjopen.com/). A chimeric forecast is also included: an equally weighted combination of human judgment consensus forecasts and of computational models from the COVID-19 Forecasthub ensemble.

    Each month, individuals contribute forecasts for incident cases, deaths, hospitalizations, cumulative number of vaccines, and variant prevalence at the US national level. Individuals can produce forecasts at the beginning of each month until the middle of the month. All forecasting ends at 2PM EST.

    Four models, 12 forecasts, 45 predictions
    Docs Example Project
    cornell A template project for learning how to interact with Zoltar. Typically, a full description of the project would go here. You could include narrative details about what seasons are included, what group has provided data, whether the project focuses on real-time or retrospective forecasts.
    One model, one forecast, 29 predictions
    Election Forecasts
    nick This project stores forecasts from multiple election forecast sites, including FiveThirtyEight and the Economist, for ease of access and comparison.
    Seven models, 945 forecasts, 76,224 predictions
    CDC Real-time Forecasts
    nick Guidelines and forecasts for a collaborative U.S. influenza forecasting project.
    44 models, 2167 forecasts, 333,713 predictions
    CDC Retrospective Forecasts
    nick Guidelines and forecasts for a collaborative U.S. influenza forecasting project.
    27 models, 5647 forecasts, 869,638 predictions