cpr (version 0.4.1)

trimmed_quantile: Trimmed Quantiles

Description

For data \(X = x_1, x_2, \ldots, x_n\), with order statistics \(x_{(1)}, x_{(2)}, \ldots, x_{(r)}\) return the quantiles for a trimmed data set, e.g., \(\boldsymbol{X} \backslash \{x_{(1)}, x_{(r)}\}\) (trim = 1), or \(\boldsymbol{X} \backslash \{x_{(1)}, x_{(2)}, x_{(r-1)}, x_{(r)}\}\) (trim = 2).

Usage

trimmed_quantile(x, trim = 1L, use_unique = TRUE, ...)

Value

a numeric vector, the return from quantile

Arguments

x

a numeric vector

trim

defaults to 1, omitting the min and the max

use_unique

logical, if true (defaults), base the quantiles on unique values, if false, base the quantiles on all data, after trimming.

...

other arguments to pass to stats::quantile

See Also

quantile

Examples

trimmed_quantile(1:100, prob = 1:23 / 24, name = FALSE)

# Warning
# trimmed_quantile(1:100, trim = .3, prob = 1:23 / 24, name = FALSE)

# no warning
trimmed_quantile(1:100, trim = 3, prob = 1:23 / 24, name = FALSE)