What people are saying - Write a review
We haven't found any reviews in the usual places.
Linear Regression Diagnostics
Various types of influential observations marked by
Influential Sets by Cooks Distance
19 other sections not shown
algorithm Andrews and Pregibon Andrews-Pregibon statistic ART4.DAT ART7.DAT artificial data sets aspects of influence Belsley bootstrap Box-Cox transformation Brant consider Cook and Wang Cook and Weisberg Cook's distance cutoff set Ri cutoff value Daniel and Wood defined deletion method denote design matrix detect influential observations detecting influential sets DFFITS distribution element of H empirical distribution function F distribution Figure fitted values given in Table Global Distance goodness-of-fit Hence high leverage points homoscedastic individual distance influence measure jackknife estimate jackknife method Lemma linear regression masking effect mg I liter Moore's Data multivariate linear model Nelder outliers plot regression diagnostics replacement method representing the influence residual and leverage response variables robust robust statistics set of influential set Ri cutoff suggested swamping phenomenon th component th diagonal element th observation tion vations vec(B vector version of Cook's Welsch xm matrix y■th