Pemodelan Principal Component Regression dengan Software R

Authors

  • Margaretha Ohyver Bina Nusantara University

DOI:

https://doi.org/10.21512/comtech.v3i1.2400

Keywords:

multicollinear, principal component regression, R software.

Abstract

Principal Component Regression (PCR) is one method to handle multicollinear problems. PCR produces principal components that have a VIF less than ten. The purpose for this research is to obtained PCR model using R software. The result is a model of PCR with two principal components and determination coefficients R(square) = 97,27%.

Dimensions

Plum Analytics

References

Jollife, I. T. (2002). Principal Component Analysis (2nd ed). New York: Springer-Verlag.

Ohyver, M. (2010). Penerapan partial least squares pada data gingerol. ComTech, 1(1): 39-47.

Pradipta, N. (2009). Metode Regresi Ridge untuk Mengatasi Model Regresi Linier Berganda yang Mengandung Multikolinearitas. Skripsi tidak diterbitkan. Universitas Sumatera Utara, Medan.

Diakses dari http://repository.usu.ac.id/bitstream/123456789/14037/1/09E01589.pdf.

Suhartono. (2008). Analisis Data Statistik Dengan R. Yogyakarta: Graha Ilmu.

Downloads

Published

2012-06-01

Issue

Section

Articles
Abstract 915  .
PDF downloaded 527  .