Essays In Applied Econometrics

Author : Seongjin Jin
Publisher :
ISBN 13 :
Total Pages : pages
Book Rating : 4.:/5 (127 download)

DOWNLOAD NOW!

Download or read book Essays on Applied Econometrics written by Seongjin Jin and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: My dissertation, titled "Essays on Applied Econometrics," develops new methods to analyze economic data and design experiments, and applies them to labor market data. Each chapter answers a real-world question by applying a new econometric method to the data. Chapter 1 studies a policy change to a cash transfer program for the elderly in South Korea. This change doubled the maximum benefit to approximately $200 per month and expanded the target group by 10%. The main estimand is the causal effect of this change on the elderly's labor supply, such as labor force participation and hours of work. I take two approaches to identify the counterfactual outcomes: a potential outcomes analysis and a structural model analysis. I find that the program's change did not affect the labor supply on the intensive or extensive margins significantly. However, if the policymaker were to increase the benefit to about $1000, the estimated structural model implies a significant reduction in labor supply on both the intensive and extensive margins. Chapter 2 introduces a novel method to adaptively design randomized experiments. For randomized experiments with a pilot stage, or multi-stage experiments, [Hahn et al., 2011] propose an adaptive experimental design that adjusts the next stage's propensity score based on data of the previous stages. I discuss how the discretization of covariates affects the precision of the estimation of the average treatment effect (ATE) through the estimated propensity score for the next stage. Also, I propose an algorithm using the bootstrap technique to find the optimal level of discretization of covariates. Monte Carlo simulations and an application with actual data show that the suggested method performs well. Chapter 3 proposes hybrid estimation for the density of fixed censored data. This approach combines a parametrically estimated density function for the censored observations with a nonparametrically estimated density for the uncensored observations. My method is less sensitive to distributional assumptions than conventional parametric methods. I apply my method to estimate the full distributions of fixed right-censored variables, such as annual earnings and the survival time of cystic fibrosis patients.