2010
DOI: 10.1080/13504850802046989
|View full text |Cite
|
Sign up to set email alerts
|

Robust vs. classical principalcomponent analysis in the presence of outliers

Abstract: Principal Component Analysis (PCA) is a very versatile technique for dimension reduction in multivariate data. Classical PCA is very sensitive to outliers and can lead to misleading conclusions in the presence of outliers. This article studies the merits of robust PCA relative to classical PCA when outliers are present. An algorithm due to Filzmoser et al. (2006) based on a modification of the projection pursuit algorithm of Croux and Ruiz-Gazen (2005) is used for robust PCA computations for a financial data s… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2013
2013
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 13 publications
(6 citation statements)
references
References 7 publications
0
6
0
Order By: Relevance
“…The outlier sensitivity of CPCA has caused the development of new approaches [2,18]. Therefore, robust estimates of PCs are an important topic, Campbell [3] and Devlin et al [6] proposed to robustify PCA by using a robust covariance matrix such as MCD.…”
Section: Robust Pcamentioning
confidence: 99%
“…The outlier sensitivity of CPCA has caused the development of new approaches [2,18]. Therefore, robust estimates of PCs are an important topic, Campbell [3] and Devlin et al [6] proposed to robustify PCA by using a robust covariance matrix such as MCD.…”
Section: Robust Pcamentioning
confidence: 99%
“…The main concern is that the outliers may skew the first few mode directions. While there are robust algorithms that are useful in stabilizing PCA in the presence of outliers [2532], it is often effective to remove identifiable outliers or simply consider a sufficiently long trajectory for which the results are significant. Generating a large number of conformational samples and removal of outliers before the C-matrix is calculated mitigates concerns about robustness of the results.…”
Section: Introductionmentioning
confidence: 99%
“…Principal component analysis (PCA) is a dimensionality reduction approach. It converts the multidimensional data into a lower dimension for more straightforward analysis (Sapra 2010 ). Traditional PCA is sensitive to anomalies and can mislead in results in the existence of anomalies.…”
Section: Condition-based Maintenance (Cbm)mentioning
confidence: 99%