Nisa, Khoirin (2006) PEMBANDINGAN REGRESI KOMPONEN UTAMA DAN ALGORITMA PROYEKSI DALAM MENGATASI MULTIKOLINEARITAS. In: Seminar Program Pengembangan Diri (PPD) 2006 Bidang MIPA.

This is the latest version of this item.

[img]
Preview
Text
Nisa PPD2006.pdf

Download (4MB) | Preview

Abstract

Principal component regression (PCR) is often used in regression with multicollinearity, but it is not a method which directly maximizes coefficient of determination (R2) since the first k principal components (PCs) with the largest variances do not always the variables that are most highly correlated with the response variable. To maximize the R2 we have to select the PCs by a procedure such as the stepwise procedure. Projection algorithm is a method which can be used to overcome multicollinearity and can also maximize the R2. In this study we aimed to compare the R2 resulted by the PCR, Stepwise-PCR and the projection algorithm by Monte Carlo simulation. We generated data from normal distribution with 1900 replications. The result shows that projection algorithm and Stepwise-PCR yield almost the same R2 and each of the two methods performs better than the PCR.

Item Type: Conference or Workshop Item (Speech)
Subjects: H Social Sciences > HA Statistics
Depositing User: DR. KHOIRIN NISA
Date Deposited: 27 Mar 2018 04:46
Last Modified: 27 Mar 2018 04:46
URI: http://repository.lppm.unila.ac.id/id/eprint/6606

Available Versions of this Item

  • PEMBANDINGAN REGRESI KOMPONEN UTAMA DAN ALGORITMA PROYEKSI DALAM MENGATASI MULTIKOLINEARITAS. (deposited 27 Mar 2018 04:46) [Currently Displayed]

Actions (login required)

View Item View Item