Publication: Inference on semiparametric models with discrete regressors
Loading...
Identifiers
Publication date
1993-02
Defense date
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
We study statistical properties of coefficient estimates of the partially linear regression model when some or all regressors, in the unknown part of the model, are discrete. The method does not require smoothing in the discrete variables. Unlike when there are continuous regressors. when all regressors are discrete independence between regressors and regression errors is not required. We also give some guidance on how to implement the estimate when there are both continuous and discrete regressors in the unknown part of the model. Weights employed in this paper seem straightforwardly applicable to other semiparametric problems.
Description
Keywords
Semiparametric partially linear model, Discrete regressors, Empirical conditional expectation estimate, Semiparametric efficiency bound, Higher order kernels