Robust Statistics: Theory and Methods

by Ricardo A. Maronna, Douglas R. Martin, Victor J. Yohai

Classical statistical techniques fail to cope well with deviations from a standard distribution. Robust statistical methods take into account these deviations while estimating the parameters of parametric models, thus increasing the accuracy of the inference. Research into robust methods is flourishing, with new methods being developed and different applications considered. Robust Statistics sets out to explain the use of robust methods and their theoretical justification. It provides an up-to-date overview of the theory and practical application of the robust statistical methods in regression, multivariate analysis, generalized linear models and time series. This unique Robust Statistics aims to stimulate the use of robust methods as a powerful tool to increase the reliability and accuracy of statistical modelling and data analysis. It is ideal for researchers, practitioners and graduate students of statistics, electrical, chemical and biochemical engineering, and computer vision. There is also much to benefit researchers from other sciences, such as biotechnology, who need to use robust statistical methods in their work.

436 pages, Hardcover

First published June 13, 2006

Reviews

[Joe · Rating 4 out of 5]

This little-known classic is a tale of international espionage, harrowing rescues, and a world threatened with annihilation by a mysterious clan known only as the "non-Gaussian outliers." Admittedly, the characters are hastily drawn, sacrificed in favor of exposition, but such an exposition!

Did you know (Spoiler alert!) that there exist location estimators nearly as efficient for normal distributions as the arithmetic mean, and certainly more efficient than the sample median, yet far more robust against outliers? Well, neither did I. But now I am a convert to some of these techniques. If you ever lie awake at night worrying about the fragility of the standard deviation as a dispersion estimator, then this is the book for you.

[Brian · Rating 5 out of 5]

I can't imagine a better book on the subject. Every time I think I have a better way to think of one of the topics covered here, or a better order in which to introduce them, I come back to this volume and realize I was wrong. If the authors want to improve it, they will not do so by altering what is there but rather, perhaps, by adding new research or some chapters on Bayesian methods.

[Somayeh]

I need to read book.