1- University of Kashan , mehdi_shams1357@yahoo.com
2- payame Noor University
Abstract: (601 Views)
In the case of large amounts of information, the data should be summarized in such a way that no valuable information is lost. Sufficient statistics are used for this purpose. In this paper, by defining similar statistics, a method for calculating these statistics is presented. Informative statistics are then defined under several different loss functions. The concept of sufficiency is used to identify two characteristics, similarity and informativeness. Finally, under several different loss functions, in the case of Gaussian observation errors and linear systems, informative statistics are found.
Type of Study:
Original Manuscript |
Subject:
stat Received: 2023/05/7 | Revised: 2024/08/17 | Accepted: 2023/08/19 | Published: 2023/12/31 | ePublished: 2023/12/31