ESTIMATION OF POPULATION DENSITY BASED ON LINE TRANSECT DATA WITH AND WITHOUT THE SHOULDER CONDITION

Abstract

In this paper, new nonparametric estimators for the population density D using line transect sampling are proposed and studied. One estimator is developed when the shoulder condition is assumed to be true and another one when this assumption is violated. The mathematical and numerical properties of these estimators are investigated and compared -via simulation techniquewith other existing estimators. A new technique is suggested to combine the proposed estimators. This technique relies on testing the shoulder condition, which in turn produces two new estimators. The simulation results show the good potential performances of the different proposed estimators.