Big data are often contaminated by outliers and heavy-tailed errors. To address this challenge, we propose the adaptive Huber regression for robust estimation and inference. The key observation is that the robustification parameter should adapt to sample size, dimension and moments for optimal tradeoff between biases and robustness. ...
Date : 12 décembre 2016
Heure : 14h30 à 15h30
Lieu : Pavillon André-Aisenstadt
Salle : 6214