引用本文:
【打印本页】   【下载PDF全文】   View/Add Comment  Download reader   Close
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 1550次   下载 1136 本文二维码信息
码上扫一扫!
分享到: 微信 更多
分布混合分析的深入研究及其在南极磷虾种群结构中的应用
鲁北伟, 王荣
中国科学院海洋研究所 青岛 266071
摘要:
介绍分布混合分析的理论,并运用分布混合分析估计南极磷虾种群体长分布概率密度函数的参数,首次给出体长取样间隔有效性和方差σ2下限的数学分析。概率密度函数式为(此处公式见PDF文件),式中x是随机变量体长的取值;K是年龄组数;Pj是第Aj年龄组的强度;qxujσj)是均值为uj;方差为σ2j的正态分布。Pjujδjj=1,2,…,K是待估计的参数。求解这些参数的最大似然估计值时,发现最速下降法是一种非常实用的方法,很容易选取Pjujσj的一组适定的初始近似解,使得迭代过程收敛。
关键词:  体长频数分布  分布混合分析  种群结构  南极磷虾
DOI:
分类号:
基金项目:国家八五攻关专题,85-905-02-01号
FURTHER STUDY ON DISTRIBUTION MIXTURE ANALYSIS AND ITS APPLICATION TO AGE-GROUPS ANALYSIS OF ANTARCTIC KRILL (EUPHAUSIA SUPERBA)
Lu Beiwei, Wang Rong
Institute of Oceanology, Chinese Academy of Sciences, Qingdao 266071
Abstract:
This paper presents the principle and algorithm of distribution mixture analysis, and its application to age-groups analysis of Antarctic krill (Euphausia superba), and presents, for the first time, the analytically deduced effective range of sampling interval and the lower limit of variance σ2j. The probability density function is (See PDF), where x is the observed value of random variable, body length; K is number of age-group; Pj is strength of the Aj age-group, q (x ; ujσj) is normal probability distribution function with mean uj and variance of σ2j, Pjujσj, j = 1,2,.., K are parameters to be estimated. The steepest descent method was used to estimate the parameters. The result shows that the steepest descent method is practical. Its deduction process is simple and easy to understand, its calculation is simple, and selection of a set of proper iterative initial values of the parameters is easy also.

A age-groups analysis of Antarctic krill by means of the steepest descent method is given in this paper. Using this method, better understanding of the population structure of Antarctic krill is obtained based on the length-frequency data (Wang et al., 1995).

Key words:  Length-frequency distribution, Distribution mixture analysis, Population structure, Antarctic krill
Copyright ©  Editorial Office for Oceanologia et Limnologia Sinica    Copyright©2008 All Rights Reserved
Supervised by: China Association for Science and Technology   Sponsored by: Chinese Society for Oceanology and Limnology, Institute of Oceanology and Limnology, CAS.
Address: 7 Nanhai Road, Qingdao, China.    Postcode: 266071    Tel: 0532-82898753  E-mail: liuxiujuan@qdio.ac.cn  
Technical support: Beijing E-Tiller Co.,Ltd.