Skip to contents

Reband any discrete distribution

Source: R/reband-discrete-data.R
reband_discrete.Rd

e.g. age banded population, or a discrete probability distribution e.g. a serial interval distribution.

Usage

reband_discrete(
  x,
  y,
  xout,
  xlim = c(0, NA),
  ytotal = c(0, sum(y)),
  digits = 0,
  labelling = c("positive_integer", "inclusive", "exclusive"),
  sep = "-"
)

Arguments

x

a set of upper limits of bands, e.g. for age: 0-14;15-64;65-79;80+ is 15,65,80,NA

y

a set of quantities for each band e.g. population figures

xout

a set of new upper limits

xlim

Upper and lower limits for x. if the last band is e.g 80+ in the input and we want to know the 85+ band in the output some kind of maximum upper limit is needed to interpolate to.

ytotal

upper and lower limits for y. If the interpolation values fall outside of x then the in and max limits of y are given by this.

digits

if the xout value is continuous then how many significant figures to put in the labels

labelling

are the xout values interpretable as an inclusive upper limit, or an exclusive upper limit, or as an upper limit of an `positive_integer`` quantity

sep

seperator for names e.g. 18-24 or 18 to 24

Value

a rebanded set of discrete values, guaranteed to sum to the same as y

Examples

ul = stringr::str_extract(england_demographics$class, "_([0-9]+)",group = 1) %>%
  as.numeric()

tmp = reband_discrete(
  ul, england_demographics$population,
  c(5,10,15,40,80), xlim=c(0,120))

tmp
#>      0-4      5-9    10-14    15-39    40-79      80+ 
#>  3745688  3361511  3384582 18173104 25360084  2464731 

sum(tmp)
#> [1] 56489700
sum(england_demographics$population)
#> [1] 56489700