Research

"Partially Linear Models with Endogeneity: a conditional moment based approach", joint with Bertille Antoine, accepted at The Econometrics Journal, Volume 25, Issue 1, January 2022, Pages 256–275.

We propose an estimator for the slope parameter in partially linear models using a smooth minimum distance approach and apply it to the impact of electrification on employment growth in South Africa, obtaining more precise estimates than previous studies.

R code for RSMD and an illustrative example

"Factor IV Estimation in Conditional Moment Models with an application to Inflation Dynamics" joint with Bertille Antoine, Journal of Financial Econometrics, April 2024.

This paper develops a new estimator that directly exploits factor-based conditional moment restrictions without the need for prior parametrization, with an empirical application estimating the New Keynesian Phillips curve using US data, revealing the equal importance of forward- and backward-looking behaviours in inflation dynamics.

R code for FSMD and an illustrated example

"Conditional Moment Restriction Approach for Panel Data: Revisiting the Democracy-Growth Relationship"

Abstract: This paper introduces a new estimator for studying the relationship between democracy and economic growth in a panel data setting. Our approach utilizes conditional moment restrictions directly, based on the foundational work of Bierens (1982), eliminating the need for parametric assumptions about the relationship between the instrument and democracy. This method reduces model specification biases and enhances estimation efficiency. Extending Integrated Conditional Moment (ICM) methods to panel data, our estimator is consistent and asymptotically normal under regularity assumptions. Applied to data from Acemoglu et al. (2019), our method produces estimates of democracy's positive impact on log GDP per capita, similar to 2SLS results but with lower standard errors. This yields more statistically significant estimates of long-term gains from democracy.

"Estimation of Heterogeneous Treatment Effects Using a Conditional Moment-Based Approach"

Submitted

This paper presents a new estimator for heterogeneous treatment effects in a partially linear model, integrating a Robinson transformation, the Smooth Minimum Distance (SMD) method, and regularized model selection techniques like Lasso. It is applied to estimate Medicaid's effects from the Oregon Health Insurance Experiment, yielding more reliable results than traditional GMM approaches.

"Partially Identified Heterogeneous Treatment Effect with Selection: an Application to Gender Gaps" joint with 

Donald Poskitt and Xueyan Zhao

Abstract: This paper addresses the sample selection model within the context of the gender gap problem, where even random treatment assignment is affected by selection bias. By offering a robust alternative free from distributional or specification assumptions, we bound the treatment effect under the sample selection model with an exclusion restriction, an assumption whose validity is tested in the literature. This exclusion restriction allows for further segmentation of the population into distinct types based on observed and unobserved characteristics. For each type, we derive the proportions and bound the gender gap accordingly. Notably, trends in type proportions and gender gap bounds reveal an increasing proportion of always-working individuals over time, alongside variations in bounds, including a general decline across time and consistently higher bounds for those in high-potential wage groups. Further analysis, considering additional assumptions, highlights persistent gender gaps for some types, while other types exhibit differing or inconclusive trends. This underscores the necessity of separating individuals by type to understand the heterogeneous nature of the gender gap.

"Information-based LASSO to select instruments in conditional moment models", joint with Bertille Antoine