../../_images/logo.png

NetworkDistribution#

class fracability.Statistics.NetworkDistribution(parent, obj: rv_continuous = None, parameters: tuple = None, fit_data: NetworkData = None)#

Class used to represent a fracture or fracture network length distribution. It is essentially a wrapper for the scipy rv distributions.

property AD_distance: float#

Calcuate the Koziol and Green distance between the empirical and the fitted model :return: Float. The KG distance

Notes

Kim 2019, Tests based on EDF statistics for randomly censored normal distributions when parameters are unknown

property AD_rank#
property AIC: float#

Property that returns the classic Akaike Information Criterion (1974) of the distribution :return:

property AICc: float#

Property that returns the Akaike Information Criterion (for small number of values) of the distribution :return:

property Akaike_rank#
property BIC: float#

Property that returns the Bayesian Information Criterion of the distribution :return:

property KG_distance: float#

Calcuate the Koziol and Green distance between the empirical and the fitted model :return: Float. The KG distance

Notes

Kim 2019, Tests based on EDF statistics for randomly censored normal distributions when parameters are unknown

property KG_rank#
property KS_distance: float#

Calcuate the Kolmogorov-Smirnov distance between the empirical and the fitted model :return: Float. The KS distance

Notes

Kim 2019, Tests based on EDF statistics for randomly censored normal distributions when parameters are unknown

property KS_rank#
property Mean_rank#
property b5: float#

Property that returns the 5th percentile of the distribution :return:

property b95: float#

Property that returns the 95th percentile of the distribution :return:

cdf(x_values: array = None)#
property distribution: rv_continuous#

Property that returns or sets a frozen ss.rv_continuous class :return:

property distribution_name: str#

Property that returns the name of the given distribution :return:

property distribution_parameters: tuple#

Property that returns the parameters of the frozen distribution :return:

log_pdf(x_values: array = None) array#

Property that returns the logpdf calculated on the data :return:

log_sf(x_values: array = None) array#

Property that returns the logsf calculated on the data :return:

property max_log_likelihood: float#

Property that returns the log likelihood of the distribution. The likelihood is calculated by adding the cumulative sum of the log pdf and log sf of the fitted distribution. :return:

property mean: float#

Property that returns the mean of the frozen distribution :return:

property median: float#

Property that returns the median of the distribution :return:

property mode: list#

Property that returns the mode(s) of the pdf of the given distribution :return:

property n_distribution_parameters: int#

Property that returns the number of parameters of the frozen distribution. the args method returns the shape parameters. This means that except for normal and logistic the loc counts as a parameter. To fix this we subtract -1 to all except normal and logistic. :return:

property std: float#

Property that returns the standard deviation of the distribution :return:

property var: float#

Property that returns the variance of the distribution :return: