Tsallis distribution

In statistics, a Tsallis distribution is a probability distribution derived from the maximization of the Tsallis entropy under appropriate constraints. There are several different families of Tsallis distributions, yet different sources may reference an individual family as "the Tsallis distribution". The q-Gaussian is a generalization of the Gaussian in the same way that Tsallis entropy is a generalization of standard Boltzmann–Gibbs entropy or Shannon entropy.[1] Similarly, if the domain of the variable is constrained to be positive in the maximum entropy procedure, the q-exponential distribution is derived.

The Tsallis distributions have been applied to problems in the fields of statistical mechanics, geology, anatomy, astronomy, economics, finance, and machine learning. The distributions are often used for their heavy tails.

Note that Tsallis distributions are obtained as Box-Cox transformation[2] over usual distributions, with deformation parameter \lambda=1-q. This deformation transforms exponentials into q-exponentials.

Procedure

In a similar procedure to how the normal distribution can be derived using the standard Boltzmann–Gibbs entropy or Shannon entropy, the q-Gaussian can be derived from a maximization of the Tsallis entropy subject to the appropriate constraints.

Common Tsallis distributions

q-Gaussian

See q-Gaussian.

q-exponential distribution

See q-exponential distribution

q-Weibull distribution

See q-Weibull distribution

See also

Notes

  1. Tsallis, C. (2009) "Nonadditive entropy and nonextensive statistical mechanics-an overview after 20 years", Braz. J. Phys, 39, 337–356
  2. Box, George E. P.; Cox, D. R. (1964). "An analysis of transformations". Journal of the Royal Statistical Society, Series B 26 (2): 211–252. JSTOR 2984418. MR 192611.

Further reading

External links

This article is issued from Wikipedia - version of the Friday, March 25, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.