scientific article; zbMATH DE number 4003303

From MaRDI portal
Publication:4728057





[For the entire collection see Zbl 0615.00010.] A regression is constructed using prior knowledge, data, models, and a fitting (estimation) process of some form. It is important to know when the resulting regression depends heavily on a small part of the prior knowledge, on a small part of the data, or on the exact choice of model or fitting process. We want models that are sensitive to the issues of interest, but perform well (i.e., are less sensitive) when assumptions are violated. In this paper we present an overview of the theory, application, and computation of regression diagnostics, especially those related to the analysis of influential data.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4728057)