[Ml-stat-talks] Fwd: Wilks Statistics Seminar: Aurore Delaigle, Today, April 25 @ 12:30pm, Sherrerd Hall 101

Barbara Engelhardt bee at princeton.edu
Mon Apr 25 09:05:35 EDT 2016


Talk of interest.

---------- Forwarded message ----------
***   Wilks Statistics Seminar   ***

DATE:  Today, April 25, 2016

TIME:   12:30pm

LOCATION:   Sherrerd Hall 101

SPEAKER:   Aurore Delaigle, University of Melbourne

TITLE:  Methodology for Deconvolution When the Error Distribution Is Unknown

ABSTRACT:   In nonparametric deconvolution problems, in order to estimate
consistently a density or distribution from a sample of data contaminated
by additive random noise it is often assumed that the noise distribution is
completely known or that an additional sample of replicated or validation
data is available. Methods have also been suggested for estimating the
scale of the error distribution, but they require somewhat restrictive
smoothness assumptions on the signal distribution, which can be hard to
verify in practice. Taking a completely new approach to the problem, we
argue that data rarely come from a simple, regular distribution, and that
this can be exploited to estimate the signal distributions using a simple
procedure, often giving very good performance. Our method can be extended
to other problems involving errors-in-variables, such as nonparametric
regression estimation. Its performance in practice is remarkably good,
often equalling (even unexpectedly) the performance of techniques that use
additional data to estimate the unknown error distribution. This is joint
work with Peter Hall.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20160425/1640bbb6/attachment.html>


More information about the Ml-stat-talks mailing list