한국해양대학교

Detailed Information

Metadata Downloads

Subset Selection in a Multiple Linear Regression Model: An Improved Tabu Search

Title
Subset Selection in a Multiple Linear Regression Model: An Improved Tabu Search
Alternative Title
다중선형회귀모델의 변수선택을 위한 개선된 타부서치 알고리즘
Author(s)
김정태
Issued Date
2016
Publisher
한국해양대학교 대학원
URI
http://kmou.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000002235191
http://repository.kmou.ac.kr/handle/2014.oak/8475
Abstract
This thesis deals with the subset selection that is a vital combinatorial optimization problem in multivariate statistics. It is the selection of the optimal subset of variables in order to reliably construct a multiple linear regression model. Since this problem has NP-complete nature, the larger the size of the variables, the harder to find the optimal solution. In general, many metaheuristic methods have been developed to tackle the problem. In the subset selection problem, two typical metaheuristics, which are tabu search and hybrid GSA (genetic and simulated annealing algorithm), was proposed. However, they have some shortcomings, that is, the tabu search takes a lot of computing time due to many neighborhood moves and GSA’s solution quality is less accurate. This paper proposes an improved tabu search algorithm to reduce moves of the neighborhood and adopt the appropriate move search strategy. To evaluate the performance of the proposed method, a comparative study is performed on both the literature data sets and simulation data sets. Computational results show that the proposed method outperforms the previous metaheuristics in terms of the computing time and solution quality.
Appears in Collections:
데이터정보학과 > Thesis
Files in This Item:
000002235191.pdf Download

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

Browse