*Published in UNSRI Journal of Science Research Vol.17, No.2, May 2015 *

19Reviews , 10Followers

UPI’s FPMIPA Department of Mathematical Education

dewirachmatin@upi.edu

** **

**Essence:** ** A **new method for reducing high-dimensional space developed from the PCA method *Weighted Principal Component Analysis* (WPCA) was introduced by J.F. Pinto da Costa, H. Alonso and L. Roque (2011). Therefore in the PCA method, pearson correlation coefficient is very sensitive to the presence of interference and pencilan, then in this WPCA method used a new correlation coefficient involving the rank of each observation for each variable. To provide an overview of this WPCA method, in this article the program for WPCA was created with S-PLUS 2000 software applied to infant data (Damayanti,2008) as well as runner time record data (Johnson, 2007), and the results are compared to the results of the classic PCA method.

**Keywords :** *Rank, Correlation* *Coefficient, PCA and WPCA.*