Abstract
This talk explores two related ideas. The first one is to let the data guide us in downweighting observations in order to improve the fit of the model. Such procedures can, for example, be used to identify subsets of the data that closely fit a particular regression model and as such provide an alternative to robust methods. The second idea consists in fragmenting the data into pieces, analyzing the pieces separately and constructing a global analysis by putting the fragments back together. In analyzing a two-way table, for example, different interaction patterns may apply to different parts of the table.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fox, P.N. (1996). The CIMMYT Wheat Program’s International Multienvironment Trials. Plant Adaptation and Crop Improvement. M. Cooper and G.L. Hammer, eds. Oxford: CAB International.
Morgenthaler, S. and Schumacher, M. (1999). Robust Analysis of a Response Surface Design. Chemometrics and Intelligent Laboratory Systems, Amsterdam: Elsevier.
Mosteller, F. and Tukey, J.W. (1977). Data Analysis and Regression. Reading: Addison-Wesley.
Tukey, J.W. (1949). One Degree of Freedom for Non-Additivity. Biometrics.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Morgenthaler, S. (2002). Weights and Fragments. In: Härdle, W., Rönz, B. (eds) Compstat. Physica, Heidelberg. https://doi.org/10.1007/978-3-642-57489-4_98
Download citation
DOI: https://doi.org/10.1007/978-3-642-57489-4_98
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-7908-1517-7
Online ISBN: 978-3-642-57489-4
eBook Packages: Springer Book Archive