quantileCI {MKmisc}

Confidence Intervals for Quantiles

Description

These functions can be used to compute confidence intervals for quantiles (including median).

Usage

quantileCI(x, prob = 0.5, conf.level = 0.95, method = "exact",
           minLength = FALSE, na.rm = FALSE)
medianCI(x, conf.level = 0.95, method = "exact",
         minLength = FALSE, na.rm = FALSE)
madCI(x, conf.level = 0.95, method = "exact", minLength = FALSE,
      na.rm = FALSE, constant = 1.4826)

Arguments

x

numeric data vector

prob

quantile

conf.level

confidence level

method

character string specifing which method to use; see details.

minLength

logical, see details

na.rm

logical, remove NA values.

constant

scale factor (see mad).

Details

The exact confidence interval (method = "exact") is computed using binomial probabilities; see Section 6.8.1 in Sachs and Hedderich (2009). If the result is not unique, i.e. there is more than one interval with coverage proability closest to conf.level, then a matrix of confidence intervals is returned. If minLength = TRUE, an exact confidence interval with minimum length is returned.

The asymptotic confidence interval (method = "asymptotic") is based on the normal approximation of the binomial distribution; see Section 6.8.1 in Sachs and Hedderich (2009).

Value

A list with components

estimate

the sample quantile.

CI

a confidence interval for the sample quantile.

Author(s)

Matthias Kohl [e-mail]

References

L. Sachs and J. Hedderich (2009). Angewandte Statistik. Springer.

See Also

binom.test, binconf

Examples

## To get a non-trivial exact confidence interval for the median
## one needs at least 6 observations
set.seed(123)
x <- rnorm(8)
## exact confidence interval not unique
medianCI(x)
madCI(x)

## minimum length exact confidence interval
medianCI(x, minLength = TRUE)
madCI(x, minLength = TRUE)

## asymptotic confidence interval
medianCI(x, method = "asymptotic")
madCI(x, method = "asymptotic")

## confidence interval for quantiles
quantileCI(x, prob = 0.4)
quantileCI(x, prob = 0.6)