Nisa, Khoirin and Herawati, Netti (2010) ANALISIS DISKRIMINAN LINIER ROBUST MENGGUNAKAN METODE VOLUME ELLIPSOID MINIMUM. PROSIDING SEMINAR NASIONAL SAINS MIPA DAN APLIKASINYA 2010 (SN SMAP 10).

This is the latest version of this item.

[img]
Preview
Text
Nisa SMAP 2010.pdf

Download (533kB) | Preview

Abstract

Discriminant Analysis is a technique concerned with separating distinct sets of objects or observations into groups and allocating new objects into previously defined groups. In the classical approach, discriminant rules are often based on the empirical mean and covariance matrix of the data. But these estimates can be sensitive to outlying data, and they become inappropriate at contaminated data sets. In this paper, we consider the problem of robust Linear Discriminant Analysis (LDA) by inserting robust estimates of mean and covariance matrix using Minimum Volome Ellipsoids (MVE) estimator. We measure the performance of classical and robust LDA through the total probability of misclassification or the Error Rate as a performance criterion. This paper shows that LDA based on MVE can improve the error rate of discriminant analysis.

Item Type: Article
Subjects: H Social Sciences > HA Statistics
Q Science > QA Mathematics
Divisions: Fakultas Matematika dan Ilmu Pengetahuan Alam (FMIPA) > Prodi Matematika
Depositing User: DR. KHOIRIN NISA
Date Deposited: 27 Mar 2018 04:45
Last Modified: 27 Mar 2018 04:45
URI: http://repository.lppm.unila.ac.id/id/eprint/6619

Available Versions of this Item

Actions (login required)

View Item View Item