Robust Principal Component Analysis Based on Modified Minimum Covariance Determinant in the Presence of Outliers
Principal component analysis (PCA) is not resistant to outliers existing in multivariate data sets. The results which are obtained by using classical PCA are far from real values in the presence of outliers. Therefore, using robust versions of PCA is favorable. The easiest way to obtain robust princ...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Istanbul University
2016-09-01
|
| Series: | Alphanumeric Journal |
| Subjects: | |
| Online Access: |
http://alphanumericjournal.com/media/Issue/volume-4-issue-2-2016/aykiri-gozlemlerin-varliginda-uyarlanmis-en-kucuk-kovaryans-_5YcrjpT.pdf
|
| Summary: | Principal component analysis (PCA) is not resistant to outliers existing in multivariate data sets. The results which are obtained by using classical PCA are far from real values in the presence of outliers. Therefore, using robust versions of PCA is favorable. The easiest way to obtain robust principal components is to replace classical estimates of the location and scale parameters with their robust versions. Robust estimations of location and scale parameters can be found with minimum covariance determinant (MCD) providing high breakdown point. In this study, algorithm of MCD is modified using Jackknife resampling approach and results of this modification are examined. Proposed robust principal component analysis (RPCA) based on modified MCD (MMCD) method that is modified using Jaccknife resampling are evaluated over two real data with different outlier ratios. In the light of obtained results, it can be said that RPCA based on MMCD is better than RPCA based on MCD in the presence of outliers. |
|---|---|
| ISSN: | 2148-2225 2148-2225 |