What happens if in the principal component analysis the Pearsonian is replaced by the Brownian coefficient of correlation?

Mishra, Sudhanshu K (2014): What happens if in the principal component analysis the Pearsonian is replaced by the Brownian coefficient of correlation?

Abstract

The Brownian correlation has been recently introduced by Székely et al. (2007; 2009), which has an attractive property that when it is zero, it guarantees independence. This paper investigates into the effects and advantages, if any, of replacement of the Pearsonian coefficient of correlation (r) by the Brownian coefficient of correlation (say, ρ), other things remaining the same. Such a replacement and analysis of its effects have been made by the Host-Parasite Co-evolutionary algorithm of global optimization applied on six datasets.

• Item Type: MPRA Paper
• Language: English
• Keywords: Brownian correlation, Principal Component Analysis, Global Optimization, Host-Parasite Co-evolutionary algorithm, Iris Flower Dataset, 1985 Auto Imports Database, Levy distribution, outliers
• Subjects: C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C13 - Estimation: General; C - Mathematical and Quantitative Methods > C4 - Econometric and Statistical Methods: Special Topics > C43 - Index Numbers and Aggregation; C - Mathematical and Quantitative Methods > C4 - Econometric and Statistical Methods: Special Topics > C45 - Neural Networks and Related Topics; C - Mathematical and Quantitative Methods > C6 - Mathematical Methods ; Programming Models ; Mathematical and Simulation Modeling > C61 - Optimization Techniques ; Programming Models ; Dynamic Analysis; C - Mathematical and Quantitative Methods > C6 - Mathematical Methods ; Programming Models ; Mathematical and Simulation Modeling > C63 - Computational Techniques ; Simulation Modeling; C - Mathematical and Quantitative Methods > C8 - Data Collection and Data Estimation Methodology ; Computer Programs > C87 - Econometric Software

References

Anderson, E. (1936). The species problem in Iris. Annals of the Missouri Botanical Garden. 23(3):457-509.

Donald Bren School of Information and Computer Sciences (1996). Machine-learning-databases/autos. https://archive.ics.uci.edu/ml/machine-learning-databases/autos/. University of California, Irvine.

Eckerle, K. and NIST (1979). Circular Interference Transmittance Study. National Institute of Standards and Technology, US Dept. of Commerce. http://www.itl.nist.gov/div898/strd/nls/data/eckerle4.shtml.

Fisher, R.A. (1936). The use of multiple measurements in taxonomic problems. Annals of Eugenics. 7(2): 179-188.

Hotelling, H. (1933). Analysis of a Complex of Statistical Variables into Principal Components. Journal of Educational Psychology. 24(6): 417–441.

Kibler, D., Aha, D.W. and Albert, M. (1989). Instance-based prediction of real-valued attributes. Computational Intelligence. 5(2): 51-57.

Mishra, S.K. (2013). Global Optimization of Some Difficult Benchmark Functions by Host-Parasite Coevolutionary Algorithm. Economics Bulletin. 33(1): 1-18.

National Crime Records Bureau (?). Incidence And Rate Of Violent Crimes During 2011. Ministry of Home Affairs, Govt. of India. http://ncrb.nic.in/CD-CII2011/cii-2011/Table%203.1.pdf

Székely, G.J. and Rizzo, M.L. (2009). Brownian distance covariance. The Annals of Applied Statistics. 3(4): 1236-1265.

Székely, G.J., Rizzo, M.L. and Bakirov, N. K. (2007). Measuring and testing independence by correlation of distances. Ann. Statist. 35(6): 2769-2794.