Back

QJHong Model for Novel Coronavirus Disease 2019 (COVID-19) in the United States

Mehta, A.; Hong, Q.-J.

2023-07-31 epidemiology
10.1101/2023.07.30.23293233 medRxiv
Show abstract

We present the methodology of the QJHong model, a machine learning predictive model we built to forecast the COVID-19 daily cases, number of daily deaths, fatality rate, reproductive number, and overall trends in the United States (both national and individual states). We measure the accuracy and compare it to other predictive models. Several forecast analyses consistently demonstrate that the QJHong model outperforms other models submitted to the COVID-19 Forecast Hub with regards to forecasting national data. The Forecast Hub is utilized by the Centers for Disease Control and Prevention (CDC) as a means of disseminating official public communications pertaining to the ongoing pandemic. As such, our model has been identified as a premier performer within this context.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.