On q-Generalized Extreme Values under Power Normalization with Properties, Estimation Methods and Applications to Covid-19 Data
Accepted: April 2022
Keywords:
extreme value theory, survival discretization approach, entropy, estimation methods, simulation, COVID-19Abstract
In this paper, the q-analogues of the generalized extreme value and its discrete version under power normalization, are introduced. The additional parameter q allowing for increased modeling exibility. The continuous extended model can yield several types of hazard rate functions, and their supports can be finite and infinite as well as bounded above or below. Furthermore, the new models are capable of modelling skewed data, especially which have very extreme observations. Some statistical properties of the proposed continuous distribution are provided. The model parameters are estimated by utilizing various estimation approaches. A simulation study is performed to test the performance of the estimators based on different sample sizes. Finally, three distinctive real data sets are analyzed to show the exibility of the proposed model.
