Plot categorical forecasts — plot_forecast

This function creates a bar plot for categorical forecasts. See examples for demonstration of usage.

Usage

plot_forecast_categorical(categorical_forecast, format = "hubverse")

Arguments

categorical_forecast: Either a tibble with categorical forecasts created with forecast_categorical or prepared forecast submission in "hubverse" format (see Details)
format: Either "hubverse" or "legacy"; the "hubverse" format will require an input forecast that includes output for "pmf" (see Details); default is "hubverse"

Value

A ggplot2 object with categorical forecasts shown as a stacked bar plot.

Details

The categorical plotting function works both with "legacy" formatting (i.e., format used in the 2022-23 FluSight season) and the "hubverse" formatting (i.e., format used in the 2023-24 FluSight season). Unlike the "legacy" format, the "hubverse" format allows for quantile and categorical forecasts to be co-mingled in the same submission object. If the format is specified as "hubverse", then the plot_forecast_categorical() function will interally look for the "pmf" forecasts.

Examples

if (FALSE) { # \dontrun{
# Retrieve hospitalization data
h_raw <- get_hdgov_hosp(limitcols=TRUE)
# Prepare and summarize hospitalization data to weekly resolution
prepped_hosp <- prep_hdgov_hosp(h_raw)
# Create a keyed time series tibble with only locations of interest
prepped_tsibble <- make_tsibble(prepped_hosp,
                                     epiyear = epiyear,
                                     epiweek=epiweek,
                                     key=location)
# Run with default constrained ARIMA, nonseasonal ETS, no NNETAR
hosp_fitfor <- ts_fit_forecast(prepped_tsibble,
                               horizon=4L,
                               outcome="flu.admits")
# Prepare forecast for quantile submission format
prepped_forecast <- format_for_submission(hosp_fitfor$tsfor, method = "ts")
# Run categorical summary of quantiles for the time series ensemble
categorical_forecast <- forecast_categorical(prepped_forecast$ensemble, prepped_hosp)
# Plot the categorical forecast
plot_forecast_categorical(categorical_forecast, format = "legacy")
} # }