Abstract:
We consider the problem of estimating coverage intervals (both one-sided and two-sided) of the standard two-sided power distribution (STSP-distribution) based on sample data. We check the quality of the obtained estimates using the Monte Carlo method. We study the properties of the maximum likelihood estimates of the parameters of the original distribution and estimate the influence of their bias on the quality of estimating coverage intervals. We also give examples demonstrating that the obtained estimates can be used for continuous distributions that can be approximated by a family of STSP-distributions.
Keywords:standard two-sided power distribution, STSP-distribution, coverage interval, Monte Carlo method, maximum likelihood estimate.