A position for a post-doctoral researcher in the area of robust
estimation is available at
IDSIA (Lugano, Switzerland, www.idsia.ch). Occupancy degree is 100%.
In performing a Bayesian analysis concerning the unknown value of a
parameter, one is required to express the prior beliefs about the
parameter in the form of a prior distribution. However, prior
information is often imprecise and cannot be expressed with a single
prior distribution. In this case, an alternative approach is to
express prior information in terms of a family consisting of all prior
distributions that are deemed reasonable in representing one’s prior
beliefs. Inferences should then carried out by considering the whole
family of distributions. This approach is known as Bayesian robustness
- examples are contamination models, intervals of measures, density
ratio classes, etc. In the case almost no prior information is
available on the parameter, the family of distributions should be as
large (i.e., as least-committal, or as weak) as possible in order to
describe this state of prior ignorance. In this respect, the class of
families known as near-ignorance priors has recently been the focus of
much research: these are models that express a minimal state of
beliefs a priori, thus representing a condition of prior ignorance on
some random quantities of interest, while always leading to
informative posterior inferences. Prior near-ignorance models can be
regarded as an objective-minded approach to inference, in that prior
beliefs are maximally weakened in favor of information coming from
You can obtain further information and references from our website:
**Keywords: robust estimation, Bayesian robustness, set of
distributions, imprecise probability, consistency of the posterior
distribution, asymptotic analysis.
The goal of this project is first of all to develop new methods for
robust estimation based on families of distributions with particular
emphasis on near-ignorance models. Therefore the project will be
mostly based initially on methodological developments at the
theoretical level. The second part of the project will focus on the
application of those methods to problems in which robustness is an
issue such as, for instance, regression, filtering, time series
analysis (e.g., robust Kalman filter) etc. The project is granted by
the Swiss National Science Foundation and focuses on basic research.
-have excellent mathematical skills, preferable in the area of
probability and statistics,
-have both a Master and a Ph.D. degree in mathematics or statistics or
physics or engineering or related area,
-have a strong publication record on topics related to statistical
estimation, robust estimation, applied probability.
-strong commitment to research and publication.
-a two-years position (degree of occupancy 100%), with possibility of
-international working environment (English is the official language);
-funded travels in case of papers accepted by well-known international
Applicants should submit the following documents, written in English:
-list of exams and grades obtained during the Bachelor and the Master
-list of three references (with e-mail addresses);
-brief statement on how their research interests fit the topics above
-publication list and possibly links to the thesis.
Applications should be submitted by 31 August 2012 through the online
form at the address:
Incomplete applications, or submitted to other addresses, or beyond
the deadline will be not accepted.
For further information please refer to the official page for this position: