compareMedians: Permutation test for difference in medians of unpaired...
In dkneis/knut: Kneis utilities

Description Usage Arguments Value Note Author(s) References See Also Examples

Tests whether the medians of two samples are equal.

1 2	compareMedians(x1, x2, alternative = "two.sided", n = 2000, name1 = "median1", name2 = "median2")

`x1`	First sample (numeric vector).
`x2`	Second sample (numeric vector).
`alternative`	One of 'two.sided', 'less', or 'greater'.
`n`	Number of replicates (integer).
`name1`	Name of 1st sample to be used in output (character string).
`name2`	Name of 2nd sample to be used in output (character string).

Numeric vector of length 3. The first two elements hold the empirical sample medians. The third element, named 'p', holds the p-value representing the probability that the two medians are equal.

See example below.

David Kneis david.kneis@tu-dresden.de

Based on the code at https://stat.ethz.ch/pipermail/r-help/2010-May/238562.html.

No linkes.

x <- c( 18,1.2,2.44,2.35,4.68,7.5 )
y <- c( 40.2,11.4,5.78,11.5,9.13,15.3,12.5 )
print(compareMedians(x, y, alternative="less"))
print(compareMedians(x, y, alternative="two.sided"))
print(compareMedians(x, y, alternative="greater"))
## Not run: 
library("rcompanion")
print(percentileTest(x=x, y=y, test="median", r=2000))

## End(Not run)