My research interests are in the fields of Econometrics, Social Networks and Applied Microeconomics. In particular, I am interested in methods for networks. In my Job market paper, I develop a peer effects models for count data. Peer effects on count data are generally estimated using a linear-in-means model which assumes that the dependent variable is continuous. I show that the counting nature of the dependent variable is important and assuming it is continuous underestimates the peer effects. The model can be used to estimate peer effects on the number of occurrences of an event in a constant period (e.g., the number of cigarettes smoked, the number of times people eat fast food, go to the gym, etc.). In another paper, a joint work with Vincent Boucher, we propose a new method for estimating peer effects by releasing the common assumption that the network is entirely observed or measured without errors. This is better suited to networks constructed from survey data. Lastly, one of my papers (Accepted in Journal of Financial Econometrics) with Arnaud Dufays and Alain Coen, uses a variables selection method to detect relevant changes in the parameters of linear-in-mean autoregressive (AR and ARX) models for time series subject to multiple structural breaks.
- Selective Linear Segmentation for Detecting Relevant Parameter Changes (with Aunaud Dufays and Alain Coën), 2020, Accepted in Journal of Financial Econometrics [Paper]
- Count Data Model with Social interactions, 2020 (Job Market Paper) - [Paper] [R Package] [Package vignettes]
I present a model of peer effects in which the dependent variable takes integer values. I present an incomplete information game rationalizing the model, and I provide sufficient conditions under which the equilibrium of the game is unique. I estimate the model's parameters using the Nested Partial Likelihood method. I show that the counting nature of the dependent variable is important and that assuming incorrectly that it is continuous significantly underestimates the peer effects. I estimate peer effects on the the number of extracurricular activities in which students are enrolled. Increasing the number of activities in which friends are enrolled by one implies an increase of 0.295 in the number of activities in which students are enrolled, when controlling for network endogeneity. Ignoring the endogeneity of the network overestimates the peer effects.
- Estimating Peer Effects using Partial Network Data (with Vincent Boucher), 2020 - [Paper] [R Package]
We study the estimation of peer effects through social networks when researchers do not observe the network structure. Instead, we assume that researchers know (have a consistent estimate of) the distribution of the network. We show that this assumption is sufficient for the estimation of peer effects using a linear-in-means model. We present and discuss important examples where our methodology can be applied. In particular, we provide an empirical application to the study of peer effects on students’ academic achievement.
Work In Progress
- Network in Repo Auctions
- Hospital Length of Stays, Readmissions and Deaths: A Multiple Spell Survival Analysis (whith Damien Echevin and Bernard Fortin)
- R package for estimating Count Data Model with Social interactions (available on GitHub)
- R package for estimating Peer Effects using Partial Network Data, joint with Vincent Boucher (available on GitHub)
- Fast sampling from von Mises Fisher distribution - [link]
- Estimate spatial econometric models with SAS - [link]