Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/119306
Title: Selang kepercayaan untuk koefisien B1 dari garis regresi apabila ragam galat tidak homogen dengan metode ols, wls dan bootstrap
Authors: Mattjik, Ahmad Ansori
Aunuddin, Aunuddin
Fauzy, Akhmad
Issue Date: 1998
Publisher: IPB University
Abstract: Slope estimation of a regression line and its confidence interval can be constructed by OLS (Ordinary Least Squares) and WLS (Weighted Least Squares) method. Confidence interval being found by both method must satisfy assumption about normality of error term with zero mean and common variance for OLS method and heteroscedasticity for WLS method and independency of error term. NC (Nanayakkara and Cressie) method is a method aimed to reduce variance of estimate resulted from OLS method when error term has non- constant variance.
URI: http://repository.ipb.ac.id/handle/123456789/119306
Appears in Collections:MT - Mathematics and Natural Science

Files in This Item:
File Description SizeFormat 
1998afa.pdf
  Restricted Access
Fullteks8.64 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.