Parametric and noparametric approaches were used to fit line transect data. Different parametric detection functions are suggested to compute the smoothing parameter of the nonparametric fourth-order kernel estimator. Among the different candidate parametric detection functions, the researcher suggests to use Akaike Information Criterion (AIC) to select the most appropriate one of them to fit line transect data. More specifically, four different parametric models are considered in this research. Where as two models were taken to satisfy the shoulder condition assumption, the other two do not. Once the appropriate model is determined, it can be used to select the smoothing parameter of the nonparametric fourth-order kernel estimator. As the researcher expected, this technique leads to improve the performances of the fourth-order kernel estimator. For a wide range of target densities, a simulation study is performed to study the properties of the proposed estimators which show the superiority of the resulting proposed fourth-order kernel estimator over the classical kernel estimator in most considered cases.
Digital Object Identifier (DOI)
Algarni, Ali and Almutlg, Ahmad
"Akaike Information Criterion and Fourth-Order Kernel Method for Line Transect Sampling (LTS),"
Applied Mathematics & Information Sciences: Vol. 10:
1, Article 27.
Available at: https://digitalcommons.aaru.edu.jo/amis/vol10/iss1/27