Skip to contents

This function evaluates whether or not the evaluated signal interval tapers (i.e., decreases in width) as horizons progress. The interval used in this plausibility component is drawn from the upper and lower bounds of the forecasted prediction interval. As such, the only accepted signal format is forecast, which will include upper and lower bounds.

Usage

plane_taper(location, input, seed)

Arguments

location

Character vector with location code; the location must appear in input and seed

input

Input signal data to be scored; object must be one of forecast

seed

Prepared seed

Value

A list with the following values:

  • indicator: Logical as to whether or not the prediction interval width tapers with advancing horizons

  • widths: Consecutive interval widths for forecasted data

Examples


## read in example observed data and prep observed signal
hosp <- read.csv(system.file("extdata/observed/hdgov_hosp_weekly.csv", package = "rplanes"))
hosp$date <- as.Date(hosp$date, format = "%Y-%m-%d")
prepped_observed <- to_signal(hosp, outcome = "flu.admits", type = "observed", resolution = "weeks")

## read in example forecast and prep forecast signal
fp <- system.file("extdata/forecast/2022-10-31-SigSci-TSENS.csv", package = "rplanes")
prepped_forecast <- read_forecast(fp) %>%
  to_signal(., outcome = "flu.admits", type = "forecast", horizon = 4)

## prepare seed with cut date
prepped_seed <- plane_seed(prepped_observed, cut_date = "2022-10-29")

## run plane component
plane_taper(location = "19", input = prepped_forecast, seed = prepped_seed)
plane_taper(location = "44", input = prepped_forecast, seed = prepped_seed)