Least Trimmed Squares Asymptotics

Seminar abstract

Asymptotic theory for the Least Trimmed Squares (LTS) estimator is derived. We consider a linear multiple regression model that allows for outliers both in the error term and in the regressors including leverage points. We show that the LTS estimator is bounded, consistent and asymptotically normal in this model. The asymptotic variance does not require the commonly used correction factors.

How to join this seminar

This seminar is free to attend with no need to register in advance.

We welcome you to join this seminar on Wednesday 4 May 2022 at 2pm.

Speaker bio

Dr Vanessa Berenguer-Rico 

Vanessa studied Economics at University of Barcelona and has a PhD in Economics from University Carlos III de Madrid. After her PhD, she moved to the University of Oxford: first as a Postdoctoral Research Fellow in Nuffield college, then as an Associate Professor in Mansfield college and the Department of Economics.


She is passionate about Econometrics and Statistics. She has worked on non-linear, non-stationary models and more recently has started to work on robust statistics methods and models for outliers.