Abstract:
In this paper we construct two new goodness-of-fit tests for Pareto I type distribution family with an arbitrary shape-parameter $\lambda$, based on the new received characterization. We describe their limiting distributions, calculate local Bahadur efficiencies under close alternatives and perform asymptotic comparison of our test statistics.
Key words and phrases:characterization, Pareto I type distribution, $U$-statistics, Bahadur efficiency, large deviations, Kullback–Liebler information.