<COVID-19 Edition>
Epidemiological Forecasting by DELPHI
This system, originally called ‘Epicast’, was designed to forecast flu-like illnesses. We are now using it to forecast COVID-19, the flu-like illness now in global pandemic. To help with our COVID-19 forecasting effort, please sign up below.
Research Overview

In this research study, being conducted by Carnegie Mellon University, we ask you each Friday during the COVID-19 pandemic to predict current and future COVID-19 activity within a group of states or within one or more Health and Human Services regions of the United States. You will be able to enter and submit your predictions online using any modern web browser, and we expect that the entire process will take no more than two minutes per region. There are no expected risks or benefits to participants in this study.

Your privacy and the confidentiality of your predictions will be strictly protected; we will not share your email address or your individual predictions without your prior written consent. The study is entirely voluntary, and you are free to stop participating or withdraw entirely at any time.

The primary contact is Brian Clark (Email:, and the principal investigator is Roni Rosenfeld (Email:; Office: GHC 8002). If you have questions pertaining to your rights as a research participant, or to report concerns with this study, you should contact the Carnegie Mellon University Office of Research Integrity and Compliance (Email:

New User Signup
I have read and understood the above:
I am 18 years old or older:
Your email address:
Your nickname (optional):
  Email My User ID