Customer Reviews

Robust Statistics (Wiley Series in Probability and Statistics)

5 star:		(2)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

In the 1970s Peter Huber was one of the innovative geniuses that developed the area of robust statistical methods. After the famous Princeton robustness study that Huber participated in there was a scattered set of techniques that were shown to be robust estimators of location based on simulations over wide classes of probability distributions. Huber and Hampel were the leaders at putting together some mathematical theory for robustness.

This book was the first attempt to unify the mathematical ideas into a general theory. It is intended for research statisticians and is a masterpiece for the subject. There are now other good books of a more practical nature. Huber also wrote a nice monograph in the SIAM series around the same time. It is now 20 years since the publication of the book and it perhaps deserves to be recognized by republication as a Wiley Classic.

If you need clear explanations about robust statistics, if you need ideas to perform robust regression, or if you need some ground to develop robust algorithms, all you need is this text, and only this text. It covers theoretical as well as practical aspects of robust statistics. If you need more modern theoretical materials on robust statistics, Rieder's Asymptotic Robust Statistics is the companion text.