Robust Regression

If you have data that kind-of lies on a line but there are nasty outliers, try some variation on medians. A good book is:
• Rousseeuw, P.J. and Leroy, A.M. (1987), Robust Regression and Outlier Detection, Wiley-Interscience, New York (Series in Applied Probability and Statistics), 329 pages. ISBN 0-471-85233-3.

  Rousseeuw's home page is http://win-www.uia.ac.be/u/rousse/.

  Software for the "least median of squares" algorithm is called PROGRESS. It is at: Statlib under the name PROGRESS.

  If you don't use UNIX, take the file apart in your editor.