bin {descstat}

Bin series

Description

A new class called bin is provided, along with different functions which enable to deal easily with bins, ie creating bin objects (as_bin) coercing bins to numerical values (as_numeric), merging bins (cut) and checking than an object is a bin (is_bin).

Usage

as_bin(x)

is_bin(x)

as_numeric(x, pos = 0, xfirst = NULL, xlast = NULL, wlast = NULL)

## S3 method for class 'bin'
cut(x, breaks = NULL, ...)

## S3 method for class 'character'
cut(x, breaks = NULL, ...)

## S3 method for class 'factor'
cut(x, breaks = NULL, ...)

## S3 method for class 'character'
extract(data, ..., .name_repair = "check_unique")

## S3 method for class 'factor'
extract(data, ..., .name_repair = "check_unique")

Arguments

x

a character or a factor: the first and last characters should be any of [, (, ⁠]⁠, ⁠)⁠ and the other characters should be interpreted as two numerical values separated by a comma,

pos

a numeric between 0 and 1, 0 for the lower bond, 1 for the upper bond, 0.5 for the center of the class (or any other value between 0 and 1), which indicates to as_numeric how the bins should be coerced to numerical values,

xfirst, xlast

the center of the first (last) class, if one wants to specify something different from the average of the lower and the upper bonds,

wlast

in the case where the upper bond is infinite and xlast is not provided, the width of the last class is set to the one of the before last class. If wlast is provided, it is set to the width of the before last class times wlast,

breaks

a numerical vector of breaks which should be a subset of the initial set of breaks. If only one break is provided, all the bins with greater values are merged,

...

see base::cut() for the cut method and tidyr::extract() for the extract method,

data

a character or a factor containing bins,

.name_repair

see tidyr::extract().

Details

• extract methods for characters and factors are provided which split the character strings in a four tibble columns: the open bracket, the lower bound, the upper bound and the closing bracket.

• as_bin takes as argument a character or a factor that represents a bin, check the consistency of the string and return a bin object with levels in the correct order and NAs when the strings are malformed,

• the default cut method takes a numerical series as argument and returns a factor containing bins according to a break vector; for the bin's method, the break should be a subset of the original set of breaks and a bin with fewer levels results,

• as_numeric converts a bin to a value of the underlying variable defined by its relative position (from 0 lower bound to 1 upper bound in the bin),

• is_bin check if the argument is a bin.

Value

as_bin returns a bin object, is_bin a logical, the extract method a tibble, as_numeric a numeric and the cut method a bin object with fewer levels.

Author(s)

Yves Croissant

Examples

# create a factor containing bins using cut on a numeric
z <- c(1, 5, 10, 12, 4, 9, 8)
bin1 <- cut(z, breaks = c(1, 8, 12, Inf), right = FALSE)
# extract the elements of the levels in a tibble
extract(bin1)
# coerce to a bin object
bin2 <- as_bin(bin1)
# coerce to a numeric using the center of the bins
as_numeric(bin2, pos = 0.5)
# special values for the center of the first and of the last bin
as_numeric(bin2, pos = 0.5, xfirst = 5, xlast = 16)
# same, but indicating that the width of the last class should be
# twice the one of the before last
as_numeric(bin2, pos = 0.5, xfirst = 5, wlast = 2)
# merge in order to get only two bins
cut(bin2, breaks = c(1, 12))
# if length of breaks is 1, this is the value for which all the bins
# containing greater values are merged
cut(bin2, breaks = 8)
# check that bin1 and bin2 are objects of class bin
is_bin(bin1)
is_bin(bin2)