A Simulation Study on Local Likelihood Methods for Abundance Estimation from Presence/Absence Maps

• Main Authors: William Gomez, 戈枚司
• Other Authors: Wen-Han Hwang
• Format: Others
• Language: en_US
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/cgi-bin/gs32/gsweb.cgi/login?o=dnclcdr&s=id=%22107NCHU5337003%22.&searchmode=basic

碩士 === 國立中興大學 === 統計學研究所 === 107 === Estimating abundance based on a presence/absence map is an interesting and important issue in many research areas. This study adopts the kernel smoothing technique and uses the information around each cell of the presence/absence map to estimate the abundance of the single cell. Based on a random Poisson model and a paired negative binomial model, we develop the corresponding local constant likelihood estimation method. The paper uses simulation studies to examine the effects of estimators and the effects of selected bandwidth values used in the local likelihood methods. We also use a real data set to demonstrate the use of the proposed method.