We consider m distributions in which the first m - 1 are obtained by multip
licative exponential distortions of the mth distribution, which is a refere
nce. The combined data from m samples, one from each distribution, are used
in the semiparametric large-sample problem of estimating each distortion a
nd the reference distribution and testing the hypothesis that the distribut
ions are identical. The approach generalizes the classical normal-based one
-way analysis of variance in the sense that it obviates the need for a comp
letely specified parametric model. An advantage is that the probability den
sity of the reference distribution is estimated from the combined data and
not only fi om the,,mth sample. A power comparison with the t and F tests a
nd with two nonparametric tests, obtained by means of a simulation, points
to the merit of the present approach. The method is applied to rain-rate da
ta from meteorological instruments.