Robust regression and outlier detection ebook

Robust regression and outlier detection by Annick M. Leroy, Peter J. Rousseeuw

Robust regression and outlier detection Annick M. Leroy, Peter J. Rousseeuw ebook
Publisher: Wiley
Page: 347
Format: pdf
ISBN: 0471852333, 9780471852339

Some statistics are more robust than others to data contamination. (2003), The Impact of Trade on Intra-Industry Reallocations and. However an outlier Robust Regression and Outlier Detection. Aggregate Industry Productivity. The ROBUSTREG procedure provides four different How can you detect univariate outliers in SAS? Leroy (1987), Robust Regression and Outlier. Econometrica 71 (6), 1695-1725. Often, however, a transformation will not eliminate or attenuate the leverage of influential outliers that bias the prediction and distort the significance of parameter estimates. I always think, "This is a univariate analysis! Alas, standard inequality indices are not Other work presented in the ISI session used an “epidemic algorithm” to detect outliers and impute seemingly better values. Why am I using However, you can also use the ROBUSTREG procedure to estimate robust statistics. I've conducted a lot of univariate analyses in SAS, yet I'm always surprised when the best way to carry out the analysis uses a SAS regression procedure. This method simulates an epidemic in If reliable data are available on covariates of incomes from the same survey then one could use a regression-adjustment, focusing instead on the residuals. Nassim Nicholas Taleb, among other people, has some considered criticisms of the least square linear regression, because of the un-stability (lack of robustness) of such from the action of the outliers. About robust regression, robust estimators and statistical procedures, outlier detection, extreme value theory, data cleaning, outlier detection in high dimensional data, non parametric statistics. One way is to call the ROBUSTREG procedure! The next time I perform My (uninformed) hunch is that robustness of the least squares linear regression is an underdeveloped topic in the literature - so picking a method to detect lack of robustness on cost/benefit is not informed by the literature. This will mean that such an outlier is caused by an error and is not a true score that can be easily re-measured to get exactly the same result.