This dissertation contains three essays in the field of econometric theory. The first essay focuses on single-index binary choice regression model. A new class of semiparametric estimators based on indirect inference is proposed to estimate the regression coefficients. It is demonstrated that the proposed estimation methodology is feasible under weak distributional assumptions and robust to misclassification of responses. The second essay examines the estimation of threshold regression models with dependent data. In particular, the integrated difference kernel estimator is used as a plug-in estimator which facilitates estimation of a wide array of parametric, semiparametric and nonparametric threshold regression models. The third essay studies the identification and estimation of non separable panel data models with index structure and correlated random effects. The parameter vectors of interest are shown to be identified up to scale and could be estimated by a generalized method of moments estimator with moment conditions based on average derivative and outer product of the difference of derivatives of the regression function.
|Qualification||Doctor of Philosophy|
|Award date||20 May 2019|
|Place of Publication||Tilburg|
|Print ISBNs||978 90 5668 587 4|
|Publication status||Published - 2019|