plot_cough_rate_error: Plot cough rate error as a function of monitoring time.
In hyfe-ai/hyfer: R Utilities for Interacting with Hyfe Data

How much monitoring is enough to estimate the overall cough rate accurately? This is a difficult question to answer, but this plot can help approximate it. It shows the cough rate estimate based on 1 hour of monitoring, 2 hours, 3 hours, ..., up until the entire cough record. The plot also displays the 'overall' cough rate, using the entire monitoring record, and horizontal lines indicating a range of acceptable error around the 'true' rate.

1	plot_cough_rate_error(hours, cutoff = 0.5, min_error = 0.1, toplot = TRUE)

`hours`	A dataframe of hourly cough counts. Minimum requirements: A column `coughs` with coughs counted in each hour, and a column `rec` with monitoring time during that hour (0 - 1). For example, you can use `ho$hours` (from a `hyfe` object) or the output of `simulate_cough_rate()`.
`cutoff`	The amount of monitoring within an hour in order for that hour to be included in cough rate calculations. Default is 0.5 (half hour of monitoring during the hour).
`min_error`	Minimum acceptable error. Default is 10% (0.1).
`toplot`	If `TRUE` (the default), a plot will dispaly the result.

The hours dataframe with additional columns: estimate, the cough rate estimate up to that hour of monitoring; error, the error in that estimate based upon the overall cough rate; prop_error, the proportional error.

hyfe-ai/hyfer documentation built on Dec. 20, 2021, 5:53 p.m.