RT Dissertation/Thesis
T1 Essays on duration and count data models
A1 Costa Gomes de Sant'Anna, Pedro Henrique
AB This thesis is formed by three chapters related to duration and count data models.In the first chapter, "Testing for Uncorrelated Residuals in Dynamic Count Modelswith an Application to Corporate Bankruptcy", I propose new model checks for dynamiccount models. Both portmanteau and omnibus-type tests for lack of residual autocorrelationare considered, and the resulting test statistics are asymptotically pivotal wheninnovations are uncorrelated, but possibly exhibiting higher order serial dependence.Moreover, the tests are able to detect local alternatives converging to the null at theparametric rate T −1/2, with T the sample size. I examine the finite sample performanceof the test statistics by means of a Monte Carlo experiment. Finally, using a dataset onU.S. corporate bankruptcies, I use the new goodness-of- t tests to check if different riskmodels are correctly specified.In the second chapter, "Nonparametric Tests for Conditional Treatment Effects withDuration Outcomes", I propose new nonparametric tests for treatment effects when theoutcome of interest, typically a duration, is subjected to right censoring. The new testsare based on Kaplan-Meier integrals, and do not rely on distributional assumptions,shape restrictions, nor on restricting the potential treatment effect heterogeneity acrossdifferent subpopulations. The proposed tests are consistent against fixed alternatives andcan detect nonparametric alternatives converging to the null at the parametric n􀀀1=2-rate,n being the sample size. The finite sample properties of the proposed tests are examinedby means of a Monte Carlo study. I illustrate the use of the proposed policy evaluationtools by studying the effect of labor market programs on unemployment duration basedon experimental and observational datasets. The third chapter, "A Simple GMM for Randomly Censored Data", is a joint workwith Miguel A. Delgado. This paper proposes a simple yet powerful GMM setup toestimate parametric regression models when the outcome of interest is subjected to rightcensoring. The estimation procedure is based on Kaplan-Meier integrals, and is suitablefor both linear and nonlinear models, with possible non-smooth moment conditions. Wederive general conditions for consistency and asymptotic normality of the parameters ofinterest. Finally, a small scale simulation study demonstrate satisfactory finite sampleproperties.
YR 2015
FD 2015-06
LK https://hdl.handle.net/10016/21502
UL https://hdl.handle.net/10016/21502
LA eng
NO .
NO Application to corporaty bankruptcy ; Nonparametric test for conditional treatment effects with duration outcomes ; A simple GMM for radomly censored data (with Miguel A. Delgado)
DS e-Archivo
RD 29 jun. 2024