Testing Of Symmetry Based On Cumulative Past And Residual Extropy Of Record Values

نویسندگان
doi
10.22034/jirss.2026.2082667.1175
چکیده

This paper proposes new nonparametric tests for symmetry based on cumulative past extropy and cumulative residual extropy of record values, motivated by a recent characterization of symmetric distributions by Gupta and Chaudhary ( 2024 ). The proposed estimators are inspired by the methodology introduced by Vasicek ( 1976 ). The proposed tests do not require estimation of the centre of symmetry, making them robust and easy to implement. Their asymptotic properties and consistency are established, and critical values are obtained via Monte Carlo simulations. Power is evaluated under various asymmetric alternatives. Results show that the proposed tests perform competitively and often outperform existing symmetry tests while maintaining the nominal significance level. Application of the test to six real-world datasets confirms its e ff ectiveness in detecting symmetric and asymmetric behavior through significant p-values.