metadata-UMass-ExpertCrowd.txt

team_name: UMass-Amherst Influenza Forecasting Center of Excellence
team_abbr: UMass
institution_affil: UMass-Amherst Influenza Forecasting Center of Excellence
team_funding: CDC
team_experience: FluSight challenges since 2015/2016

model_name: Expert consensus distributions
model_abbr: ExpertCrowd
model_output: 
model_repo: https://github.com/tomcm39/COVID19_expert_survey

model_contributors: thomas mcandrew <mcandrew@umass.edu> Nicholas G. Reich <nick@schoolph.umass.edu> 
Model_targets: wk ahead cumulative deaths
Target_loc: US, some states
time_horizon:
Data_format: all requested quantiles
forecast_startdate: 2020-04-13
forecast_frequency: weekly

data_inputs_known: expert elicitation
data_source_known: 

this_model_is_an_ensemble: TRUE
this_model_is_unconditional: TRUE

methods: >-
    A consensus forecast from experts in the modeling of infectious disease, public health, and epidemiology. 

methods_long: >-
    April 22: Experts in the modeling of infectious disease, public health, and epidemiology provide forecasts of 
    the smallest, most likely, and largest possible cumulative number of deaths at the US level. We transform each 
    expert's three estimates into a triangular probability distribution and take an equally-weighted average of 
    expert predictive densities.