期刊名称：Proceedings of the National Academy of Sciences
印刷版ISSN：0027-8424
电子版ISSN：1091-6490
出版年度：2021
卷号：118
期号：51
DOI：10.1073/pnas.2111453118
语种：English
出版社：The National Academy of Sciences of the United States of America
摘要：Significance
Validated forecasting methodology should be a vital element in the public health response to any fast-moving epidemic or pandemic. A widely used model for predicting the future spread of a temporal process is an autoregressive (AR) model. While basic, such an AR model (properly trained) is already competitive with the top models in operational use for COVID-19 forecasting. In this paper, we exhibit five auxiliary indicators—based on deidentified medical insurance claims, self-reported symptoms via online surveys, and COVID-related Google searches—that further improve the predictive accuracy of an AR model in COVID-19 forecasting. The most substantial gains appear to be in quiescent times; but the Google search indicator appears to also offer improvements during upswings in pandemic activity.
Short-term forecasts of traditional streams from public health reporting (such as cases, hospitalizations, and deaths) are a key input to public health decision-making during a pandemic. Since early 2020, our research group has worked with data partners to collect, curate, and make publicly available numerous real-time COVID-19 indicators, providing multiple views of pandemic activity in the United States. This paper studies the utility of five such indicators—derived from deidentified medical insurance claims, self-reported symptoms from online surveys, and COVID-related Google search activity—from a forecasting perspective. For each indicator, we ask whether its inclusion in an autoregressive (AR) model leads to improved predictive accuracy relative to the same model excluding it. Such an AR model, without external features, is already competitive with many top COVID-19 forecasting models in use today. Our analysis reveals that 1) inclusion of each of these five indicators improves on the overall predictive accuracy of the AR model; 2) predictive gains are in general most pronounced during times in which COVID cases are trending in “flat” or “down” directions; and 3) one indicator, based on Google searches, seems to be particularly helpful during “up” trends.
关键词：COVID-19; forecasting; hotspot prediction; time series; digital surveillance
