Fuzzy hypotheses tests for crisp data using non-asymptotic fuzzy estimators, fuzzy critical values and a degree of rejection or acceptance

The fuzzy hypotheses testing is based on critical values and fuzzy test statistics produced by fuzzy estimators. This approach is particularly useful in critical situations, where subtle comparisons between almost equal statistical quantities have to be made. In such cases the fuzzy hypotheses tests...

Full description

Saved in:
Bibliographic Details
Published in:Evolving systems 2021-09, Vol.12 (3), p.723-740
Main Authors: Mylonas, Nikos, Papadopoulos, Basil
Format: Article
Language:English
Subjects:
Artificial Intelligence
Complex Systems
Complexity
Engineering
Original Paper
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The fuzzy hypotheses testing is based on critical values and fuzzy test statistics produced by fuzzy estimators. This approach is particularly useful in critical situations, where subtle comparisons between almost equal statistical quantities have to be made. In such cases the fuzzy hypotheses tests give better results than the crisp ones, since they give us the possibility of partial rejection or acceptance of H 0 using a degree of rejection or acceptance obtained by ordering fuzzy numbers. In our approach we use non-asymptotic fuzzy estimators which are triangular shaped fuzzy numbers that generalize the fuzzy estimators based on confidence intervals in such a way that they are functions.
ISSN:1868-6478
1868-6486
DOI:10.1007/s12530-021-09370-9