Mapping distributions in the entropy-parametric space
Vitaly G. Polosin · Journal of Physics Conference Series · 2020
Abstract The paper discusses the possibility of approximate identification of asymmetric models in the space of information and probability signs of the sample data distributions. The paper contains a description of the features of the method for displaying asymmetric models in the entropy-parametric space of distribution signs. In particular a brief analysis of the shortcomings of the mapping of distributions in the space of parametric signs of asymmetry and kurtosis is given. For the construction a space of distribution signs, the author is proposed an entropy coefficient for asymmetric distributions that it is used as an independent interval informational estimate for observation sample data distribution. The material is illustrated by the example of a widespread family of generalized gamma distribution.