Robust Principal Component Analysis Based On Modified Minimum Covariance Determinant In The Presence Of Outliers
Journal Title: Alphanumeric Journal - Year 2016, Vol 4, Issue 2
Abstract
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.
Authors and Affiliations
B. Barış Alkan
Keywords
Related Articles
- EP ID EP197293
- DOI 10.17093/aj.2016.4.2.5000189525
- Views 148
- Downloads 0
Mobile Phones As Distracting Tools In The Classroom: College Students Perspective
There are more than six billion users of cell phones worldwide. The need for people to connect with others frequently is apparent in our society. People use mobile phones everywhere and every time, to call and text frien...
Modeling Natural Gas Prices Volatility
Researches done so far indicate that oil reserves around the word will most probably have been used up in 50 year’s time. This fact has necessitated the researches and use of new energy sources which can be alternative t...
Determining the Risk Factors Causing Cancer with Logistic Regression Analysis
The number of cancer patients is gradually increasing, and the main cause of this disease is believed widely to be genetic. However, the mere cause of this disease is not genetic. The purpose of this study was to determi...
Determination of Effective Critical Factors in Successful Efficiency Measurement of University Departments by Using Fuzzy DEMATEL Method
Successful efficiency measurement of university departments is very important issue in today’s globalised world and depends on paying high attention on critical input and output factors affecting efficiency measurement....
A Dynamic Performance Management Model
This article aims to develop an organizational performance management model in a dynamic structure, formed in the scientific methods gathering together. All the criteria, expected to be in the performance evaluation form...