Nonparametric instrumental regression with right-censored duration outcomes.

Details

Supervisors

Van Keilegom, Ingrid

Faculty

Louvain School of Management

Degree label

Master [120] : ingénieur de gestion, à finalité spécialisée

Abstract

This thesis studies the effect of cash bonuses on unemployment duration using a non-parametric framework. A discrete instrumental variable independent from the error term of the model and explaining the discrete treatment is used to tackle the endogeneity issue. The proposed model is robust to random right censoring and is applied to the Illinois Unemployment Incentive Experiments data. We estimate quantile treatment effects and show that the efficiency of cash bonuses on unemployment duration varies, depending on the recipient of the bonus (employer or employee) and on the duration that is considered (first spell of unemployment or full benefit year). We also show that the efficiency depends on the ethnicity, gender, age and earnings of an individual by proposing a way to add covariates to the model.