No Detail Found!

Renovated partial plots and hat matrix for censored regression model.

Author(s)
Dong Q. Wang School of Mathematics, Statistics & Operations Research, Victoria University of Wellington, Wellington, New Zealand
Nazrina Aziz School of Mathematics, Statistics & Operations Research, Victoria University of Wellington, Wellington, New Zealand
Lingyun Zhang School of Engineering and Advanced Technology, Massey University, Wellington, New Zealand
S. E. Ahmed Department of Mathematics & Statistics, University of Windsor, Windsor, Canada
Abstract
Partial residual plots are a useful diagnostic tool in multiple regression to clarify the effect of the ith explanatory variable on the predictive ability of the response variable when the other explanatory variables are still present in the model. The usual diagnostic plots of residuals versus explanatory variable provide standard checks on model assumptions. Partial residual plots not only do this, but have the added enhancement of including whether violations to assumptions may be adequately modeled by including curvature in the model or should be modeled by transforming responses (Aziz and Wang, 2009) This paper outlines the construction of renovated partial residuals, plots and their properties for regression with censored data.
Publication Details
Page(s) 631-645
DOI DOI not available
Published Journal: Pakistan Journal of Statistics, Volume: 25, Issue: 4, Year: 2009

Request the full text from PASTIC

0

Citations

0

Downloads

31

Views