Mathematics > Statistics Theory
[Submitted on 23 Jun 2013]
Title:Connections between Semiparametrics and Robustness
View PDFAbstract:Robust and semiparametric statistics are of the same historical origin and largely employ the same locally asymptotically normal framework. In our talk, we consider he following more intrinsic connections of both fields:
1) Robust influence curves for semiparametric models with infinite dimensional nuisance parameter; for example, for semiparametric regression (Cox), and mixture models (Neyman--Scott).
2) Adaptiveness in the sense of Stein's necessary condition of robust neighborhood models and estimators with respect to a finite dimensional nuisance parameter; for example, location, linear regression, and ARMA.
3) Semiparametric treatment of gross error deviations from an ideal model as an infinite dimensional nuisance parameter, by projection on balls; for testing, an asymptotic version of the Huber--Strassen maximin result is thus obtained.
4) Uniform and nonuniform asymptotic normality of robust and adaptive estimators, respectively, in regression and time series models.
5) Fragility of optimal one-sided tests and confidence limits obtained for convex tangent cones, by projection on cones, as opposed to stability of corresponding procedures, even two-sided, for linear tangent spaces.
6) The unknown neighborhood radius as a nuisance parameter in robustness.
The investigation relies on asymptotic techniques and on numerical evaluations of robust estimates.
The construction under 1) of robust estimates which are adaptive, an appropriate LAM estimation bound for balls under 3), and the minimization of the norm under 5) on a certain restricted set of differences of tangents from a cone are examples of challenging open problems.
Current browse context:
math.ST
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.