A simulation study on continuous biodiversity indices

• Main Authors: Tz-Yan Lu, 呂紫嫣
• Other Authors: 沈宗荏
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/09946206777092876322

碩士 === 國立中興大學 === 統計學研究所 === 99 === Simpson index (Simpson, 1949) and Shannon index (Shannon, 1948) are two most widely used diversity indexes for measuring community structures. Most of diversity indexes in the literature are developed for dealing with categorical data (e.g. species counts). However, some research interest may focus on some variables with continuous type; for example, ecologists and biologists may concern the diversity regarding body sizes/weights of fishes, DBH’s/heights of trees, etc. The present study is interested in evaluating some existing methods in the literature with such datasets. In addition, we have proposed some ways to construct the confidence intervals associated with those considered estimators. Based on the simulation study and data analysis with forest datasets, we have given some suggestions about the merits of the evaluated estimators.