EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Moh. Fandika Aqsa


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Master Theses » Statistika - S2
Posted by anis at 30/12/2006 14:26:19  •  20860 Views


STUDI PERBANDINGAN METODE REGRESI RIDGE DENGAN KUADRAT TERKECIL PARSIAL TERHADAP PENCILAN

THE COMPARISON STUDY OF RIDGE REGRESSION AND PARTIAL LEAST SQUARES METHODS WITH OUTLIER

Author :
Retnaningsih, Endang 




ABSTRAK

Kehadiran satu atau lebih pengamatan pencilan outlier dalam analisis re gresi linier berganda menimbulkan dilematis bagi para peneliti. Keputusan untuk menghilangkan pencilan tersebut hams dilandasi alasan yang kuat karena kadang- kadang dapat memberikan informasi penting yang diperlukan. Disamping pencilan pada regresi linier ganda terdapat masalah lain yaitu adanya hubungan antara dua atau lebih variabel bebasnya. Variabel-variabel bebas yang saling berkorelasi disebut multikolinearitas. Efek dari multikolinearitas yaitu tingginya koefisien determinasi tidak diikuti oleh hasil uji hipotesis yang nyata terhadap koefisien regresi dugaan. Selain itu tanda dari koefisien regresi bisa jadi tidak sama dan nilai VIF dari be- berapa variabel bebas menjadi besar. Masalah multikolinearitas ini dapat diatasi de- gan berbagai metode diantaranya metode regresi ridge ridge regression dan meto- de kuadrat terkecil parsial partial least squares. Pi dalam data yang terdapat multi- kolinearitas sering ditemui adanya pengamatan pencilan. Untuk itu diperlukan pen- deteksian pencilan dalam regresi ridge dan metode kuadrat terkecil parsial agar dipe- roleh model terbaik. Untuk mengetahui kekekaran regresi ridge dengan metode kuadrat terkecil parsial terhadap imbasan pencilan multikolinearitas maka akan dilakukan studi per- bandingan antara kedua metode tersebut. Perbandingan dilakukan dengan melihat nilai-nilai R2 PRESS serta leverage h untuk metode regresi ridge dengan berba- gai nilai tetapan bias k yang dipilih dan metode kuadrat terkecil parsial dengan ber- bagai jumlah variabel laten yang dibentuk pada data pengamatan dan data bangkitan. Berdasarkan kriteria R2 dan jumlah pencilan maka regresi ridge relatif lebih kekar dari metode kuadrat terkecil parsial tetapi bila ditinjau dari PRESS dan besar- nya nilai leverage maka metode kuadrat terkecil parsial dikatakan relatif lebih kekar terhadap pencilan dibandingkan dengan regresi ridge. Keadaan seperti ini berlaku baik untuk data pengamatan maupun data bangkitan..


ABSTRACT

The present of one or more outlier in multiple linear regression analysis causes problems for researchers. We must have strong reasons to make a decision to eliminate the outlier since sometimes it still has important information. Another problem in multiple linier regression is a relationship between two or more inde- pendent variables. Independent variables which are correlate each other are named by multicollinearity. Multicollinearity effects the high of determination coefficients that is not followed by significantly hypothesis test result of estimate regression coef- ficient. The other effects are the sign of regression coefficient can be different and high VIF value of some independent variable. This multicollinearity problems can be solved by many methods for example ridge regression and partial least squares. It is often seen outlier in the multicollinearity data thereby outlier detection is needed in ridge regression and partial least squares to gain best models. Knowing ridge regression and partial least squares robustness are calculated to the induced-outlier multicollinearity this paper compares the two methods. Com- parison is done to R2 PRESS and leverage ha values many bias constants k are choosed for ridge regression and many latent variables are used in partial least squares through observation and simulation data. Based on R2 and the number of outlier criteria ridge regression is more rela- tively robust than partial least squares whereas from PRESS and leverage value partial least squares is relatively more robust to outlier than ridge regression. This situation exist for both observation and simulation data.



Keywordsanalisis regresi linier; multikolinearitas
 
Subject:  Analisis regresi
Contributor
  1. Drs. AGUS SUHARSONO, MS
    Ir. DWIATMONO AGUS WIDODO, MIkom.
Date Create: 30/12/2006
Type: Text
Format: pdf; 54 pages
Language: Indonesian
Identifier: ITS-Master-3100002014861
Collection ID: 3100002014861
Call Number: 519.536 Ret s


Source
Teses Statistica RT 519.536 Ret s, 2001

Coverage
ITS Community

Rights
Copyright @2001 by ITS Library. This publication is protected by copyright and permission should be obtained from the ITS Library prior to any prohibited reproduction, storage in a retrievel system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. For information regarding permission(s), write to ITS Library




[ Download - Summary ]

ITS-Master-3100002014861-431.pdf




 Similar Document...




! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 3.210.201.170
using CCBot/2.0 (https://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2020 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan