LogLogistic: Log-Logistic Distribution

In windAC: Area Correction Methods

Log-Logistic Distribution

Description

The probability density function, cumulative density function, inverse cumulative density function, random generation for the log logistic distribution.

Usage

dllog(x, shape = 1, scale = 1, log = FALSE, ...)

llogSummaryStats(shape, scale)

pllog(q, shape = 1, scale = 1, lower.tail = TRUE, log.p = FALSE, ...)

qllog(p, shape = 1, scale = 1, lower.tail = TRUE, log.p = FALSE, ...)

rllog(n, shape = 1, scale = 1, ...)

Arguments

x	Vector of quantiles.
shape	Shape parameter.
scale	Scale parameter.
log	Logical; if TRUE, log densities are returned.
...	Currently ignored.
q	Vector of quantiles.
lower.tail	Logical; if TRUE (default), probabilities are P(X <= x) otherwise, P(X > x).
log.p	Logical; if TRUE, probabilities p are given as log(p).
p	Vector of probabilities.
n	Number of observations. If length(n) > 1, the length is taken to be the number required.

Details

If X is a random variable distributed according to a logistic distribution, then Y = exp(X) has a log-logistic distribution.

The log-logistic distribution with parameters shape = a and scale = s has density

f(x) = \frac{(\frac{1}{a*exp(s))})(\frac{x}{\exp{s}})^{\frac{1}{a} - 1}}{(1+(\frac{x}{\exp{s}})^{1/a})^2}

for x >= 0, a > 1, and s > 0.

The median is exp(s), mean is

\frac{a\pi*exp(s)}{sin(a*\pi)}

for 1/a > 1. The variance is

(exp(s))^2(\frac{2*\pi*a}{(sin(2*pi*a))}- \frac{(a*\pi)^2}{(sin^2(a*\pi))})

for 1/a > 2. The mode is

exp(s)(\frac{(1/a) - 1}{(1/a) + 1})^{a}

for 1/a > 1 otherwise it is zero.

Value

dllog returns vector of the densities.

pllog returns a vector of probabilities.

qllog returns a vector of quantiles.

rllog returns a vector of random log-logistic variates.

See Also

dlogis plogis qlogis rlogis

Examples

y <- rllog(5,shape=1,scale=1/3)
dllog(x=y,shape=1,scale=1/3)
dlogis(x=log(y),location=1/3,scale=1)/y

pllog(q=y,shape=1,scale=1/3)
qllog(p=seq(0,1,by=.25),shape=1,scale=1/3)

windAC documentation built on March 31, 2023, 9:30 p.m.