An Adaptation of Kernel Density Estimation for Population Abundance using Line Transect Sampling When the Shoulder Condition is Violated

Abstract

Kernel estimation is a commonly used method to estimate the population density in line transect sampling. In general, the classical kernel estimator of (0) X f , which is the probability density function at perpendicular distance x  0 , inclines to be underestimated. In this study, a power transformation of perpendicular distance is proposed for the kernel estimator when the shoulder condition is violated. The mathematical properties of the proposed estimator are derived. A simulation study is also carried out for comparing the proposed estimator with the classical kernel estimators.