Stats (experimental)#
Examples relevant to the stats
module.
Astrostatistics Tools (experimental)#
Documentation is available in the docstrings and online at https://www.astropy.org/.
Tweedie Distribution (generalized family)#
Examples relevant to the tweedie
variable.
Tweedie distributions are a very general family of distributions that includes the Gaussian, Poisson, and Gamma (among many others) as special cases.
Parameter π: The Tweedie family is defined by a parameter π, which controls the distributionβs properties. For different values of π, the distribution can take different forms.
Example of the Tweedie Family:
π=0: Gaussian (Normal) distribution
π=1: Poisson distribution
π=2: Gamma distribution
π=3: Inverse Gaussian distribution
Common Applications#
Generalized Linear Models (GLMs): The Tweedie distributions are often used in GLMs, especially in cases where the data has non-constant variance or is overdispersed (i.e., variance greater than the mean).
The Tweedie distribution is widely used in:
Insurance industry: Modeling claim amounts and policy exposure.
Medical and genomic testing: Analyzing datasets with zero-inflated and continuous positive values.
Environmental science: Rainfall modeling and hydrology studies.