Skew-Normal Distribution
Usage
dskewnorm(x, mean = 0, sd = 1, shape = 0, log = FALSE)
pskewnorm(q, mean = 0, sd = 1, shape = 0)
qskewnorm(p, mean = 0, sd = 1, shape = 0)
rskewnorm(n = 1, mean = 0, sd = 1, shape = 0)
Arguments
- x
A numeric vector of values.
- mean
A numeric vector of the means.
- sd
A non-negative numeric vector of the standard deviations.
- shape
A numeric vector of values.
- log
A flag specifying whether to return the log-transformed value.
- q
A vector of quantiles.
- p
A vector of probabilities.
- n
A non-negative whole number of the number of random samples to generate.
Value
dskewnorm
gives the density, pskewnorm
gives the distribution function, qskewnorm
gives the quantile function, and rskewnorm
generates random deviates.
pskewnorm
and qskewnorm
use the lower tail probability.
Examples
dskewnorm(x = -2:2, mean = 0, sd = 1, shape = 0.1)
#> [1] 0.04543235 0.22269638 0.39894228 0.26124507 0.06254958
dskewnorm(x = -2:2, mean = 0, sd = 1, shape = -1)
#> [1] 0.105525330 0.407161596 0.398942280 0.076779853 0.002456603
qskewnorm(p = c(0.1, 0.4), mean = 0, sd = 1, shape = 0.1)
#> [1] -1.1980898 -0.1731883
qskewnorm(p = c(0.1, 0.4), mean = 0, sd = 1, shape = -1)
#> [1] -1.6322188 -0.7540709
pskewnorm(q = -2:2, mean = 0, sd = 1, shape = 0.1)
#> [1] 0.01848493 0.13944469 0.46827448 0.82213418 0.97298466
pskewnorm(q = -2:2, mean = 0, sd = 1, shape = -1)
#> [1] 0.0449827 0.2921390 0.7500000 0.9748285 0.9994824
rskewnorm(n = 3, mean = 0, sd = 1, shape = 0.1)
#> [1] -0.436483 -1.450847 -1.282674
rskewnorm(n = 3, mean = 0, sd = 1, shape = -1)
#> [1] -1.6330964 -0.5024617 -0.1201735