Loglogistic distribution

F(x) = (x / scale)^shape / (1 + (x / scale)^shape), as actuar's dllogis() and SciPy's fisk; scale is the median. Its cdf() is Clark's loglogistic growth curve. The mean is infinite for shape <= 1 and the variance for shape <= 2; limited and layer moments always exist. Properties: d@shape, d@scale.

Usage

loglogistic_distribution(shape, scale, ptr = NULL)

Arguments

shape, scale

Finite and positive.

ptr

Internal: an existing object to wrap.

Value

A loglogistic_distribution object, which inherits from distribution.

Details

Supports the same operations as pareto.

Examples

g <- loglogistic_distribution(1.5, 24)
cdf(g, c(12, 24, 48)) # share of ultimate reported by age
#> [1] 0.2612039 0.5000000 0.7387961
lev(g, 120)
#> [1] 37.03223