cdc_aggregated_projections: Projected deaths from COVID-19 models

In seandavi/sars2pack: COVID-19 data resources and analysis tools

Projected deaths from COVID-19 models

Description

The US CDC gathers projections from several groups around the world and aggregates them into a single data resource. See the reference below for details of the models.

Usage

cdc_aggregated_projections()

Details

These models are not updated daily but more like weekly. This function will attempt to grab the latest version.

Value

a data.frame

Author(s)

Sean Davis [e-mail]

References

• https://www.cdc.gov/coronavirus/2019-ncov/covid-data/forecasting-us.html

See Also

Other data-import: acaps_government_measures_data(), acaps_secondary_impact_data(), apple_mobility_data(), beoutbreakprepared_data(), cci_us_vaccine_data(), cdc_excess_deaths(), cdc_social_vulnerability_index(), coronadatascraper_data(), coronanet_government_response_data(), cov_glue_lineage_data(), cov_glue_newick_data(), cov_glue_snp_lineage(), covidtracker_data(), descartes_mobility_data(), ecdc_data(), econ_tracker_consumer_spending, econ_tracker_employment, econ_tracker_unemp_data, economist_excess_deaths(), financial_times_excess_deaths(), google_mobility_data(), government_policy_timeline(), jhu_data(), jhu_us_data(), kff_icu_beds(), nytimes_county_data(), oecd_unemployment_data(), owid_data(), param_estimates_published(), test_and_trace_data(), us_county_geo_details(), us_county_health_rankings(), us_healthcare_capacity(), us_hospital_details(), us_state_distancing_policy(), usa_facts_data(), who_cases()

Examples

res = cdc_aggregated_projections()
head(res)
dplyr::glimpse(res)

# available models
table(res$model)

# projection targets
table(res$target)

min(res$forecast_date)
max(res$target_week_end_date)

library(dplyr)
library(ggplot2)

# FACET view
res_ny = res %>%
    dplyr::filter(location_name=='New York' & grepl('cum death', target)) %>%
    dplyr::filter(model!='UMass-MechBayes')
res_ny %>%
    dplyr::filter(location_name=='New York') %>%
    ggplot(aes(x=model, y=point, color=model)) +
    geom_errorbar(aes(ymin= quantile_0.025, ymax =  quantile_0.975)) +
    facet_wrap(facets='target_week_end_date') +
    geom_point() +
    labs(y='Projected Deaths') +
    theme_bw() +
    theme(axis.text.x=element_blank()) +
    ggtitle('Projected New York deaths for week ending')
#'
# combined view
pd <- position_dodge(width = 3) # use this to offset points and error bars
res_ny %>%
    ggplot(aes(x=target_week_end_date, y=point, color=model)) +
    geom_errorbar(aes(ymin= quantile_0.025, ymax =  quantile_0.975), position=pd) +
    geom_point(position=pd) +
    labs(y='Projected Deaths') +
    geom_line(position=pd) +
    theme_bw() +
    theme(legend.position='bottom') +
    ggtitle('Projected New York deaths for week ending')

seandavi/sars2pack documentation built on May 13, 2022, 3:41 p.m.