|| Checking for direct PDF access through Ovid
The Geo-anchored method, based on a moment-type estimator, has been developed for estimating parent population properties from a successive sample of discoveries. By substituting the expectation of the waiting time z(n+1) of the (n + 1)th discovery to occurrence for an unknown parameter λ in the anchored method, the Geo-anchored method allows estimation of inclusion probabilities directly from observed data, thus eliminating the need for a priori selection of a value of N, R, or some other feature of the parent population. Because direct estimation of N and R requires an ordered sample, the Geo-anchored method is more sensitive to the data-generating process than the anchored method. This paper presents a sensitivity study on the Geo-anchored method. The test is based on simulated discovery sequences with different assumptions regarding discovery efficiency, exploration maturity, and the shape of the parent field size distribution. As a reference for comparison, estimates from the Horvitz–Thompson estimator also are presented.