Mapping distributions in the entropy-parametric space
Mapping distributions in the entropy-parametric space
复制标题
熵参数空间中的映射分布
DOI：
10.1088/1742-6596/1515/3/032044
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
V. Polosin
中科院分区：
文献类型：
--
作者：
V. Polosin
讨论了非对称模型在信息空间近似识别的可能性和样本数据分布的概率符号。本文描述了在分布符号的熵参数空间中显示非对称模型的方法的特点。特别对不对称和峰度参数符号空间中分布映射的缺点作了简要分析。为了构造一个分布符号空间，作者提出了一个非对称分布的熵系数，并将其作为观测样本数据分布的独立区间信息估计。本文以广义伽玛分布的一个广泛族为例加以说明。
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.