Details of Research Outputs

TitleA sequential algorithm for optimization and its applications to regression analysis
Creator
Date Issued1990
Source PublicationLecture Notes in Contemporary Mathematics
ISBN7030021622
Publication PlaceBeijing, China
PublisherScience Press
Pages17-28
Abstract

In this paper, we suggest a sequential algorithm for finding the maximum/minimum of a continuous function in a multi-dimensional rectangle by number-theoretic method. The program of this method is simple, and the precision is higher than the method without using a sequential search. We can also generalize algorithm for finding the maximum of a function in some other regions such as ball, sphere and simplex. As an application, we use the algorithm to estimate the regression coefficients in some non-linear models as well as in the linear regression model with linear and nonnegative constraints.

Language英语English
KeywordNumber-theoretic method non-linear regression analysis optimization regression with constraints uniform distribution
Citation statistics
Document TypeBook chapter
Identifierhttp://repository.uic.edu.cn/handle/39GCC9TT/5339
CollectionResearch outside affiliated institution
Affiliation
1.Institute of Applied Mathematics, Academia Sinica
2.Institute of Mathematics, Academia Sinica
Recommended Citation
GB/T 7714
Fang, Kaitai,Wang, Yuan. A sequential algorithm for optimization and its applications to regression analysis. Beijing, China: Science Press, 1990: 17-28.
Files in This Item:
There are no files associated with this item.
Related Services
Usage statistics
Google Scholar
Similar articles in Google Scholar
[Fang, Kaitai]'s Articles
[Wang, Yuan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fang, Kaitai]'s Articles
[Wang, Yuan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fang, Kaitai]'s Articles
[Wang, Yuan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.