Abstract
The methods discussed in Chapters 12, 13, and 16 are very powerful if the underlying distributions are close to normal. If the distributions of errors of the points on a curve, for example, are not normal and have long tails, then, as we noted in Chapter 13, estimates of goodness of fit may be seriously biased. The tests discussed in this chapter tend to be robust, with results which are independent of the particular distribution being tested.
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.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1992 Springer Science+Business Media New York
About this chapter
Cite this chapter
Roe, B.P. (1992). Beyond Maximum Likelihood and Least Squares; Robust Methods. In: Probability and Statistics in Experimental Physics. Springer, New York, NY. https://doi.org/10.1007/978-1-4757-2186-7_17
Download citation
DOI: https://doi.org/10.1007/978-1-4757-2186-7_17
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4757-2188-1
Online ISBN: 978-1-4757-2186-7
eBook Packages: Springer Book Archive