[Forum SIS] Phd Course on Robust statistical methods (Bologna)

[Daniela Cocchi]

Università di Bologna - Dipartimento di Scienze Statistiche
Dottorato di Ricerca in Metodologia statistica per la ricerca scientifica

Phd Course on Robust statistical methods

Lecturers: Christian Hennig, UCL; Andrea Cerioli, Università di Parma; Francesca Torti, Università di Parma
Course coordinator: Angela Montanari, Università di Bologna

Monday, October 17 2011, Friday, October 21 2011, Monday October 24 2011
Time 14.30-17.30
Lecturer: Christian Hennig
Topics:
Most statistical methods are derived under model assumptions, which cannot be verified (although they can be refuted in some situations). Robust Statistics is about methodology that is more "robust", i.e., less sensitive against violation of the model assumptions.

The course will focus on simple standard situations (location and variance estimation and multiple regression) and the normal distribution assumption in order to introduce the basic principles of robust statistics.

The most problematic violation of the model assumptions in this situation are usually outliers, which occur often in real data.

The course will cover:
1) The robustness problem - how bad are outliers for classical methods?
2) Measurement of robustness - breakdown point and influence function
3) Quality of estimators - efficiency and equivariance
4) Robust estimators of location - M-estimators
5) Optimality theory for robust estimators
6) Robust estimators of the variance MAD, IQR, M-scale
7) Robust estimators for multiple regression - L1, LMS, S, MM-estimators.

References:
F. R. Hampel, E. M. Ronchetti, P. J. Rousseeuw, and W. A. Stahel. Robust Statistics the Approach Based on Influence Functions. John Wiley & Sons, New York, 1986.
P. J. Huber. Robust Statistics. John Wiley & Sons, New York, 1981. (Note that there is a recent 2nd edition, 2009, of this book, with E. Ronchetti as co-author.)
P. J. Rousseeuw and A. M. Leroy. Robust Regression and Outlier Detection. John Wiley & Sons, New York, 1987.
R. Maronna, D. Martin and V. J. Yohai. Robust Statistics: Theory and Methods.John Wiley & Sons, New York, 2006.
Friday, November 11 2011, 10-13
Lecturer: Andrea Cerioli
Topics: Inferential issues in multivariate outlier detection; robust multivariate methods and the (reweighted) MCD estimator; inference for the robust distances from the reweighted MCD estimator; multivariate outlier detection with robust distances; power improvement in outlier detection through weak control of the FWER.

Monday, November 14 2011, 10-13
Lecturer: Andrea Cerioli,
Topics: Principles of the  Forward Search for data analysis; the Forward Search for very robust regression; the Forward Search for multivariate data; inference for the robust distances from the Forward Search; multivariate outlier detection with the Forward Search; the Forward Search for cluster analysis and other data analysis problems.

Monday, November 14 2011, 14-16 (in computer room)
Lecturers: Andrea Cerioli & Dr. Francesca Torti
Topics: The FSDA toolbox for robust multivariate data analysis.

References
Books
Atkinson A.C., Riani M., (2000). Robust Diagnostic Regression Analysis, New York.
Atkinson A.C., Riani M., Cerioli A. (2004). Exploring Multivariate Data with the Forward Search. Springer, New York.
Maronna R.A., Martin R.D., Yohai V.J. (2006). Robust Statistics. Theory and Methods. Wiley, New York.
Rousseeuw P.J., Leroy A.M. (1987). Robust Regression & Outlier Detection. Wiley, New York.
Papers
Atkinson A.C., Riani M., Cerioli A. (2010). The forward search: theory and data analysis (with discussion), JKSS, 39, 117-134.
Cerioli, A. (2010). Multivariate Outlier Detection With High-Breakdown Estimators. JASA, 105, 147-156.
Cerioli, A., Farcomeni, A. (2011). Error rates for multivariate outlier detection. CSDA, 55, 544-553.
Cerioli, A., Riani M., Atkinson A.C. (2009). Controlling the size of multivariate outlier tests with the MCD estimator of scatter. StatComput, 19, 351-353.
Croux, C., Haesbroeck, G. (1999). Influence Function and Efficiency of the Minimum Covariance Determinant Scatter Matrix Estimator. JMVA, 71, 161-190.
Garcia-Escudero L.A., Gordaliza A. (2005). Generalized radius processes for elliptically contoured distributions. JASA, 100, 1036-1045.
Hardin J., Rocke D.M. (2005). The distribution of robust distances. JCGS, 14, 910-927.
Perrotta D., Riani M., Torti F. (2009). New robust dynamic plots for regression mixture detection. ADAC, 3, 263-279.
Riani M., Atkinson A.C., Cerioli A. (2009). Finding an unknown number of multivariate outliers. JRSS B, 71, 447-466.
Rousseeuw P.J., Van Driessen K. (1999). A fast algorithm for the minimum covariance determinant estimator. Technometrics, 41, 212-223.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20110930/8b78281f/attachment-0001.html>