Ph. Kvam et Fj. Samaniego, ON ESTIMATING DISTRIBUTION-FUNCTIONS USING NOMINATION SAMPLES, Journal of the American Statistical Association, 88(424), 1993, pp. 1317-1322
A nomination sample consists of independently distributed maxima from
subsamples of a population with the same underlying distribution. Nomi
nation sampling occurs when only the item with largest value is chosen
from each of n independent subsamples. If the subsamples differ in si
ze, then these observed order statistics are not identically distribut
ed, so estimation schemes built on assumptions of independently and id
entically distributed (iid) data are most likely inappropriate. But we
can exploit the structure of the data from the nomination sample by c
onditioning on the observed order of the independent maxima, and form
a least squares estimator of the distribution function that minimizes
risk with respect to squared error loss using an approach similar to o
ne found in Ferguson, where the case for iid data is presented. The re
sult is a product estimator that is consistent and compares favorably
with the nonparametric maximum likelihood estimator proposed by Boyles
and Samaniego, as indicated by graphs of mean squared error and Kolmo
gorov-Smirnov distance.