Ms. Williams et Ht. Schreuder, OUTLIER-RESISTANT ESTIMATORS FOR POISSON SAMPLING - A NOTE, Canadian journal of forest research (Print), 28(5), 1998, pp. 794-797
Poisson (3P) sampling is a commonly used method for generating estimat
es of timber volume. The usual estimator employed is the adjusted esti
mator, (Y) over cap(a). The efficiency of this estimator can be greatl
y influenced by the presence of outliers. We formalize such a realisti
c situation for high-value timber estimation for which (Y) over cap(a)
is inefficient. Here, y(i) = beta x(i) for all but a few units in a p
opulation for which y(i) is large and x(i) very small. This situation
can occur when estimating the net volume of high-value standing timber
, such as that found in the Pacific Northwest region of the United Sta
tes. A generalized regression estimator and an approximate Srivastava
estimator are not affected by such data points. Simulations on a small
population illustrate these ideas.