DocumentCode :
2982904
Title :
The Comparison of Five Discriminant Methods
Author :
Yan Hu ; Dai Yu
Author_Institution :
Statisitc Sch., Southwestern Univ. of Finance & Ecomics, Chengdu, China
fYear :
2011
fDate :
12-14 Aug. 2011
Firstpage :
1
Lastpage :
4
Abstract :
The purpose of this article is to study advantages and disadvantages about discriminant analysis with five linear methods. And through comparison,we can obtain that there are not absolute rules to tell us which is best in discriminant analysis with linear methods. The conditions in practice determine mostly the power of five methods. Through this case,we find that FDA is a most stable discriminant analysis,compared with other five methods in this case.
Keywords :
statistical analysis; FDA; discriminant functions; discriminant methods; linear discriminant analysis; stable discriminant analysis; Barium; Companies; Covariance matrix; Educational institutions; Linear discriminant analysis; Training; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Management and Service Science (MASS), 2011 International Conference on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4244-6579-8
Type :
conf
DOI :
10.1109/ICMSS.2011.5999201
Filename :
5999201
Link To Document :
بازگشت