Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/500604| Title: | Adaptive kernel estimators for population density using line transect sampling |
| Authors: | Albadareen Baker Ishaq Mohammad (P84075) |
| Supervisor: | Noriszura Ismail, Prof. Dr. |
| Keywords: | Algebra Linear Line transect Adaptive kernal estimators Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia |
| Issue Date: | 15-Jan-2021 |
| Description: | In line transect sampling, the estimation of parameter 𝑓(0), which is the probability density function of the perpendicular distances at the left boundary 𝑥=0, is important for estimating the population density. In practice, the probability density function of the perpendicular distances is unknown and is difficult to be fitted by the parametric density function. Thus, the kernel density estimation is a common method to estimate 𝑓(0). However, the classical kernel estimator of 𝑓(0) tends to have poor performance in some cases and has high negative bias with convergence rate (𝑜(ℎ2)). Based on the literature, there are several adaptations of kernel estimators that are appropriate for selected cases under some constrains. In this thesis, several adaptations of the usual kernel estimator are proposed using the logarithmic transformation, the power transformation of the perpendicular distances and the general form of Epanechnikov kernel function. The results show that the logarithmic transformations and the power transformations of the perpendicular distances are mathematically more efficient than the classical kernel estimator, whether the shoulder condition is valid or is violated. A simulation study is also carried out to compare the performance of these transformations compared to the classical kernel estimator aiming to verify the properties of these transformation using several cases of sample sizes. Simulation study is also used to find the most appropriate power transformation of the kernel estimator.,Ph.D |
| Pages: | 148 |
| Call Number: | QA184.2.A433 2021 tesis |
| Publisher: | UKM, Bangi |
| Appears in Collections: | Faculty of Science and Technology / Fakulti Sains dan Teknologi |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_128520+SOURCE1+SOURCE1.0.PDF Restricted Access | 3.34 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.