Influential observations, diagnostics and discovery tests
- 1 January 1987
- journal article
- research article
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 14 (2) , 133-142
- https://doi.org/10.1080/02664768700000017
Abstract
A Bayesian approach is presented for influential observations and predictive discordancy diagnostics. Tests of significance based on predictive distributions are devised to test the discordancy of an observation. Examples illustrating the methods are provided.Keywords
This publication has 17 references indexed in Scilit:
- A Predictive View of the Detection and Characterization of Influential Observations in Regression AnalysisJournal of the American Statistical Association, 1983
- A Predictive View of the Detection and Characterization of Influential Observations in Regression AnalysisJournal of the American Statistical Association, 1983
- Characterizations of an Empirical Influence Function for Detecting Influential Cases in RegressionTechnometrics, 1980
- Influential Observations in Linear RegressionJournal of the American Statistical Association, 1979
- Influential Observations in Linear RegressionJournal of the American Statistical Association, 1979
- The Hat Matrix in Regression and ANOVAThe American Statistician, 1978
- Detection of Influential Observation in Linear RegressionTechnometrics, 1977
- Procedures for Detecting Outlying Observations in SamplesTechnometrics, 1969
- Procedures for Detecting Outlying Observations in SamplesTechnometrics, 1969
- On Information and SufficiencyThe Annals of Mathematical Statistics, 1951