Econometrics Seminar - "Least Trimmed Squares: Nuisance Parameter Free Asymptotics" Vanessa Berenguer-Rico
UEBS Economics Seminar
Economics Seminar - Vanessa Berenguer-Rico (Oxford University)
| An UEBS Department of Economics seminar | |
|---|---|
| Speaker(s) | Vanessa Berenguer-Rico (Oxford University) |
| Date | 3 October 2025 |
| Time | 13:30 to 15:00 |
| Place | Syndicate Room B |
Event details
Abstract
Abstract: TheLeastTrimmedSquares(LTS)regressionestimatorisknowntobeveryrobustto
thepresenceof“outliers”.Itisbasedonaclearandintuitiveidea:inasampleofsize
n, it searches for the h-subsample of observations with the smallest sum of squared
residuals.Theremainingn−h observationsaredeclared“outliers”.Fastalgorithms
foritscomputationexist.Nevertheless,theexistingasymptotictheoryforLTS,based
onthetraditionalϵ-contaminationmodel,showsthattheasymptoticbehaviorofboth
regression and scale estimators depend on nuisance parameters. Using a recently
proposed new model, in which the LTS estimator is maximum likelihood, we show
that the asymptotic behavior of both the LTS regression and scale estimators are
free of nuisance parameters. Thus, with the new model as a benchmark, standard
inferenceproceduresapplywhileallowingabroadrangeofcontamination.
Location:
Syndicate Room B
