DataScience@BI seminar invites Jad Beyhum

Abstract

Abstract will come in prior to the event.

Key research areas

Jad is an econometrician working across the following two general research themes.

1. High-dimensional econometrics. Inspired by recent advances in machine learning, he develops new econometric procedures to handle high-dimensional data where the number of variables is large relative to the sample size. In particular, he studies LASSO-type procedures for duration data, factor model approaches for panel data and combining factor models and LASSO-type procedures in high-dimensional models.
2. Instrumental variables techniques. His research seeks to extend the scope of instrumental variables (IV) techniques. Specifically, he develops new IV techniques for duration data where there is a censoring, and a dynamic treatment, and enhanced approaches avoiding the first step in nonparametric IV regressions. He also works on testing and relaxing the instrumental variables assumptions.

• Read more about the speaker
• Visit our website to read more about the Department of Data Science and Analytics