The empirical moment
process in testing for the generalized two-sided power distribution
S.G. Meintanis and G. Iliopoulos
The moment process is appropriate
for analyzing random variables with a bounded support. In this paper we first
briefly discuss the properties of the empirical counterpart of this process.
Then the empirical moment process is utilized in a goodness-of-fit test for a
three-parameter version of the standard two-sided power distribution. The procedure is applied not to the original
data, but to suitably transformed data incorporating parameter estimates.
Consistency of the tests is investigated under general assumptions, and the
finite-sample behavior of the proposed method is investigated via a parametric bootstrap
procedure.
Key words and phrases: Two-sided power distribution; goodness-of-fit test; parametric boostrap; financial data.