We are continuing to curate forecasts of #COVID19 deaths in the US from multiple teams. Our current national-level ensemble combines three models and forecasts ~90K cumulative deaths by May 23 (80% PI: 82K-104K).
reichlab.io/covid19-foreca…
1/6
reichlab.io/covid19-foreca…
1/6
In general, the message of these models seems to be that #COVID19 is going to continue to exact a large toll in the US, with mortality by late May or early June likely to exceed even the upper-bound estimates of deaths caused by any recent flu season.
cdc.gov/flu/about/burd…
2/
cdc.gov/flu/about/burd…
2/
The support and vision from our collaborators at @CDCgov have been really crucial as we have built this system over the past few weeks. They are using our data for the weekly summaries hosted at their website on #COVID19 forecasting:
cdc.gov/coronavirus/20…
3/
cdc.gov/coronavirus/20…
3/
It has been more difficult than we thought to build an ensemble of these models, as different models forecast different outcomes and for different locations. Currently, we take an equal-weighted average of models that forecast 1-4 week ahead cumulative deaths for a location.
4/
4/
We have forecasts from 17 different models stored in a standardized format in our data repository. Starting to collate forecasts on incident deaths and incident hospitalizations as well. Updates on those soon!
github.com/reichlab/covid…
5/
github.com/reichlab/covid…
5/
We are indebted to all teams/individuals that have made their data publicly available, including (partial list): @meyerslab @covid_analytics @youyanggu @TAlexPerkins @alexvespi @sdelvall @sangeeta0312 @JHIDDynamics @tomcm39 @IHME_UW @SenPei_CU
6/6
6/6
