• Graduate program
  • Research
  • News
  • Events
    • Summer School
      • Climate Change
      • Gender in Society
      • Inequalities in Health and Healthcare
      • Business Data Science Summer School Program
      • Receive updates
    • Events Calendar
    • Events Archive
    • Tinbergen Institute Lectures
    • Conference: Consumer Search and Markets
    • Annual Tinbergen Institute Conference
  • Summer School
    • Climate Change
    • Gender in Society
    • Inequalities in Health and Healthcare
    • Business Data Science Summer School Program
    • Receive updates
  • Alumni
  • Magazine
Home | Events Archive | Stepwise Regression Ensembling
Seminar

Stepwise Regression Ensembling


  • Series
    Seminars Econometric Institute
  • Speaker(s)
    Stefan van Aelst (KU Leuven, Belgium)
  • Field
    Econometrics
  • Location
    Erasmus University, Polak Building, Room 2-04
    Rotterdam
  • Date and time

    September 26, 2019
    16:00 - 17:30

Abstract:

Ensembling is a powerful approach to model complex relations in high-dimensional data and yield accurate predictions. However, it is not obvious which models are best combined in an ensemble. Standard approaches are to use a predefined set of models or to use some sort of randomness to build models. We propose a data-driven approach in which the different models are grown simultaneously. The candidate variables are added to the models in a stepwise manner. Variables are combined in a single model if they work well together. Otherwise they are assigned to different models. The resulting ensemble likely overfits the data. Therefore, we regularize each of the models using lasso or elastic net penalties. These models are then combined in an ensemble to obtain the final fit. We show the performance of the method using simulations and real data examples and compare it to existing approaches.

Joint work with Anthony Cristidis and Ruben Zamar