Abstract
In this chapter the finite-sample properties of various estimators of the covariate-adjusted mean (2.25) are investigated. Estimates of (2.25) are essential ingredients for policy evaluation under the control-for-confounding-variables and under the difference-in-difference identification approaches (discussed in Section 2.1.4). Hence precise estimation of covariate-adjusted means is of importance, particularly if average treatment effects are analyzed for smaller subpopulations.1 In addition, estimation of covariate-adjusted means is also central for deriving individually optimal treatment choices in Chapter 4.
Or if the available dataset is very small.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
Notes
Some simulations were also carried out with local quadratic and local cubic matching and with a local linear variant of Hall, Park, and Turlach (1998). These estimators, however, did not perform well in small samples.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Frölich, M. (2003). Nonparametric Covariate Adjustment in Finite Samples. In: Programme Evaluation and Treatment Choice. Lecture Notes in Economics and Mathematical Systems, vol 524. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-55716-3_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-55716-3_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44328-5
Online ISBN: 978-3-642-55716-3
eBook Packages: Springer Book Archive