This PR creates additional targets to produce supplemental plots and summary tables/metrics. These will be referenced directly in the results and written into the supplement. Outlined analysis planned include but are not limited to:
[X] plot of CRPS/WIS over time stratified by horizon (CRPS for head to head comparison, WIS for hub comparison)
[X] summary of overall CRPS/WIS stratified by horizon and overall
[X] plot of average bias over time for both head-to-head and Hub analysis
[X] summary stats on the # of forecast_date-locs that ww model were more than X% (10?) better, number that were X% (10?) worse
[X] summary stats on the number of states where wastewater overall improved forecasts, number of states that overall have worse forecasts,
[X] plot of CRPS over time for both hosp only and ww model with fit and forecast comparison figure, for each forecast date and location
[X] plot of WIS over time for both hosp only and ww model + COVIDhub 4 week ensemble over time + forecast comparison + avg WIS
[X] plots of the number of sites and population coverage vs forecast performance
This PR creates additional targets to produce supplemental plots and summary tables/metrics. These will be referenced directly in the results and written into the supplement. Outlined analysis planned include but are not limited to:
See outputs here https://github.com/cdcent/cfa-forecast-renewal-ww/issues/770. I am going to open a separate issue for additional analysis discussed f2f on 2024/10/01
Closes #165 and #168