FSA (version 0.8.30)

ksTest: Kolmogorov-Smirnov Tests.

Description

Performs a one- or two-sample Kolmogorov-Smirnov test. Includes the option to perform the two-sample test using the formula notation.

Usage

ksTest(x, ...)
# S3 method for default
ksTest(
  x,
  y,
  ...,
  alternative = c("two.sided", "less", "greater"),
  exact = NULL
)
# S3 method for formula
ksTest(
  x,
  data = NULL,
  ...,
  alternative = c("two.sided", "less", "greater"),
  exact = NULL
)

Arguments

x

A numeric vector of data values or a formula (see details).

…

Parameters of the distribution specified (as a character string) by y.

y

A numeric vector of data values, a character string naming a cumulative distribution function, or an actual cumulative distribution function. See ks.test.

alternative

A string that indicates the alternative hypothesis. See ks.test.

exact

NULL or a logical that indicates whether an exact p-value should be computed. See ks.test. Not available if ties are present, nor for the one-sided two-sample case.

data

A data frame that contains the variables in the formula for x.

Value

See ks.test.

Details

This is exactly ks.test except that a formula may be used for the two-sample situation. The default version is simply a pass through to ks.test. See ks.test for more details.

See Also

ks.test.

Examples

# NOT RUN {
## see ks.test for other examples
x <- rnorm(50)
y <- runif(30)
df <- data.frame(dat=c(x,y),
                 grp=factor(rep(c("x","y"),c(50,30))),
                 stringsAsFactors=FALSE)

## one-sample (from ks.test) still works
ksTest(x+2, "pgamma", 3, 2)
ks.test(x+2, "pgamma", 3, 2)

## first two-sample example in ?ks.test
ksTest(x,y)
ks.test(x,y)

## same as above but using data.frame and formula
ksTest(dat~grp,data=df)

# }