Inference on Environmental Population Using Horvitz-Thompson Estimator

Abstract

The purpose of this paper is to propose a method to estimate the total number of the environmental population in a region. Here, we assume that the occurrence of the population in the sample site follows a distributional count pattern. We use the Horvitz-Thompson estimator for the formation of the estimation process. We provide simulated results and an example using tiger census data to illustrate the proposed estimation procedure, and compare it with estimates obtained from a commonly used method known as the simple random sampling (SRS). The results show that the mean squared error (MSE) of the proposed method is lower than that of the SRS and the statistical test also shows a good result in favour of the proposed method. Our method of estimation also shows very good outcome for smaller sample sizes compare to the SR S. As well, this method is flexible in application and also very inexpensive in both time and cost.

Keywords: detection probability, Horvitz-thompson, poisson dist ribution, environmental population, tiger census