Skip to contents

This function reads a CSV file containing weekly processed cholera data from WHO and JHU sources and downscales it to a daily time series using the MOSAIC::downscale_weekly_values method. For each country (identified by its ISO code), the function applies the downscale_weekly_values function to both the cases and deaths columns, then merges the results by date.

Usage

downscale_weekly_cholera_data(PATHS)

Arguments

PATHS

A list of file paths. This list must include an element named DATA_WHO_WEEKLY and DATA_WHO_DAILY.

Value

Invisibly returns NULL.

Details

The input CSV file cholera_country_weekly_processed.csv is expected to contain at least the following columns:

  • date_start - The start date of the week (in "YYYY-MM-DD" format, must be a Monday).

  • iso_code - The ISO code identifying the country.

  • cases - The weekly total number of cholera cases.

  • deaths - The weekly total number of cholera deaths.

The function downsamples these weekly totals to daily counts using the helper function downscale_weekly_values, which distributes any remainders symmetrically across the days of each week when integer = TRUE. The MOSAIC::convert_iso_to_country function is used to convert the ISO code to a country name.

Examples

if (FALSE) { # \dontrun{
  # Define paths list with the directory containing the weekly data file:
  PATHS <- list(DATA_WHO_WEEKLY = "/path/to/your/data")

  # Downscale the weekly cholera data to daily data:
  downscale_weekly_cholera_data(PATHS)
} # }