D'Adamo, Riccardo; (2023) Essays in econometrics. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text main_nosignatures.pdf - Other Download (2MB) | Preview

Abstract

This thesis presents new methodologies in the field of Econometrics and their application to microeconomic data. In Chapter 2, I develop a framework for estimation of optimal individualized treatment rules in the presence of partial identification. I propose an estimation procedure that ensures Neyman-orthogonality with respect to nuisance components and provide statistical guarantees for its performance. The approach is illustrated using data from the Job Partnership Training Act Study to estimate the optimal participation of workers in a job training programme. Chapter 3 presents a new instrumental variable (IV) estimator for nonlinear models with endogenous covariates. This estimator formalizes the idea that the IVs should be “excluded variables” that have no direct explanatory power for the outcome, and does not require to specify the distribution of the endogenous covariates. The theoretical properties are explored through asymptotic theory and Monte Carlo simulations, and the method is illustrated with two empirical applications. Chapter 4 develops inference methods for linear regression models with many controls and clustering. I show that commonly used cluster-robust standard errors are inconsistent when the number of controls grows proportionally with the sample size. I then propose a new standard error formula that allows to carry out valid inference in high-dimensional regression models. Monte Carlo evidence supports the theoretical results and the proposed method is illustrated with an empirical application that studies the impact of abortion on crime.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as XMLJSONCSV

Archive Staff Only

View Item