pystra.distributions.gev.GEVmax#

class GEVmax(name, mean, stdv, shape, input_type=None, startpoint=None)[source]#

Bases: Distribution

Generalized Extreme Value (GEV) distribution for maxima.

This distribution unifies the different types of extreme value distributions: Gumbel (Type I), Fréchet (Type II), and Weibull (Type III).

Arguments:
  • name (str): Name of the random variable

  • mean (float): Mean

  • stdv (float): Standard deviation

  • shape (float): Shape parameter. shape < 0.0 is Weibull, shape > 0 is Frechet.

  • input_type (any): Change meaning of mean and stdv

  • startpoint (float): Start point for seach

Raises:
  • ValueError: If shape is greater than or equal to 0.5

Notes:
  • The shape parameter shape must be less than 0.5 for finite variance.

  • shape < 0 is the Weibull case, shape = 0 is the Gumbel case, and shape > 0 is the Fréchet case.

  • This distribution is to model maxima.

Methods

cdf

Cumulative distribution function

getMean

getName

getStartPoint

getStdv

jacobian

Compute the Jacobian

pdf

Probability density function

plot

Plots the PDF of the distribution

ppf

Inverse cumulative distribution function

sample

Return a sample of the distribution of length n

setStartPoint

set_location

set_scale

u_to_x

Transformation from u to x

x_to_u

Transformation from x to u

Attributes

std_normal

cdf(x)#

Cumulative distribution function

jacobian(u, x)#

Compute the Jacobian

pdf(x)#

Probability density function

plot(ax=None, **kwargs)#

Plots the PDF of the distribution

ppf(u)#

Inverse cumulative distribution function

sample(n=1000)#

Return a sample of the distribution of length n

u_to_x(u)#

Transformation from u to x

x_to_u(x)#

Transformation from x to u