[Statlist] Research seminar in statistics March 4th 2016, GSEM University of Geneva

[Eva Cantoni]

Dear All,

We are pleased to invite you to our next Research Seminar –

Looking forward to seeing you,

Organisers :
E. Cantoni - D. La Vecchia - E. Ronchetti -
S. Sperlich - F. Trojani - M-P. Victoria-Feser


Friday March 4th, 2016 at 11h15 - Room M 5220, Uni Mail (40, bd du 
Pont-d'Arve)

Joint estimation of location and scale for robust high-dimensional 
linear regression

Po-Ling Loh, Wharton School, University of Pennsylvania

ABSTRACT:
We present recent and ongoing work on robust high-dimensional linear 
regression. In the first part of the talk, we discuss results concerning 
robust location estimation in a high-dimensional linear model, based on 
regularized M-estimation with a proper choice of loss function. Our main 
result is to show that regularized M-estimators produce statistically 
consistent solutions regardless of the distribution of the additive 
errors, provided the loss function has a bounded derivative. However, 
this result implicitly assumes that the scale of the additive error is 
known up to constant factors. We next address the problem of jointly 
estimating location and scale in a robust fashion. We develop 
statistical error bounds in settings where the covariates follow a 
sub-Gaussian distribution, and discuss extensions to settings involving 
leverage points, as well.


Visit the website: http://www.stat-center.unige.ch/ressem.html



Karen Longden Roure
Program Coordinator
MSc. in Management, MSc. in Economics, MSc. in Statistics
Université de Genève, Uni-Mail
Faculté d'Economie & Management, GSEM
40, bd. du Pont d'Arve, 1211 Genève 4
Tél:  +41.22.379.8109 (10h-14h)



-- 
Prof. Eva Cantoni
Research Center for Statistics and
  Geneva School of Economics and Management
University of Geneva, Bd du Pont d'Arve 40, CH-1211 Genève 4
gsem.unige.ch/rcs/cantoni