Professor Martin Weidner, University of Oxford & Professor Silvia Gonçalves, McGill University

Friday 18 March, 2.00pm – 5.00pm
Hybrid: Room 305, University of Glasgow and Zoom online seminar

Register at business-events@glasgow.ac.uk

Professor Martin Weidner

Biography

Martin is a Professor of Economics and a fellow of Nuffield College. He is also a Research Fellow at the Institute for Fiscal Studies and a Turing Fellow at the Alan Turing Institute. Martin is working on econometrics, with a special focus on panel data models, social networks, factor models, and high-dimensional inference.

Abstract

This paper builds on Bonhomme (2012) to develop a method to systematically construct moment conditions for dynamic panel data logit models with fixed effects. After introducing the moment conditions obtained in this way, we explore their implications for identification and estimation of the model parameters that are common to all individuals, and we find that those common model parameters are estimable at root-n rate for many more dynamic panel logit models than has been appreciated by the existing literature. In the case where the model contains one lagged variable, the moment conditions in Kitazawa (2013, 2016) are transformations of a subset of ours. A GMM estimator that is based on the moment conditions is shown to perform well in Monte Carlo simulations and in an empirical illustration to labour force participation.

Professor Silvia Goncalves

Biography

Silvia Gonçalves is a Professor of Economics at McGill University and a research member of CIREQ and CIRANO. She received her Ph.D. in 2000 from the University of California, San Diego. Before joining the Department of Economics at McGill University in 2017, she was a Professor of Economics at the University of Western Ontario (2015-2017) and at the Université de Montréal (2000-2015). Her work in econometric theory has focused on developing bootstrap methods that apply computing power to allow accurate inference fora range of statistical problems in economics, with a focus on dependent data, including financial data, panel data, and spatial data. She received the first CWEN prize for research by a young woman researcher in a Canadian University in 2010 and currently serves as a Co-Editor of the Journal of Financial Econometrics and as an Associate Editor of the Journal of Econometrics, JBES, Econometrics Journal, Journal of Time Series Analysis and the Portuguese Economic Journal.

Abstract

Many empirical studies estimate impulse response functions that depend on the state of the economy. Most of these studies rely on a variant of the local projection (LP) approach to estimate the state-dependent impulse response functions. Despite its widespread application, the asymptotic validity of the LP approach to estimating state-dependent impulse responses has not been established to date. We formally derive this result for a structural state-dependent vector autoregressive process. The model only requires the structural shock of interest to be identified. A crucial condition for the consistency of the state-dependent LP estimator of the response function is that current and future states are conditionally mean independent of the structural shocks, given the information available at the time the shock is realised. This rule out models in which the state of the economy is a function of current or future realisations of the outcome variable of interest, as is often the case in applied work. Even when the state is a function of past values of this variable only, consistency may hold only at short horisons.

Further information: Katie Allan-Mackenzie at business-events@glasgow.ac.uk

First published: 11 March 2022

<< 2022