Semiparametric estimation with generated covariates
We study a general class of semiparametric estimators when the in nite-dimensional nuisance parameters include a conditional expectation function that has been estimated nonparametrically using generated covariates. Such estimators are used frequently to e.g. estimate nonlinear models with endogenou...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Book/Monograph Working Paper |
| Language: | English |
| Published: |
Berlin
SFB 649, Economic Risk
2014
|
| Series: | SFB 649 discussion paper
2014-043 |
| In: |
SFB 649 discussion paper (2014-043)
|
| Subjects: | |
| Online Access: | Verlag, Volltext: http://sfb649.wiwi.hu-berlin.de/papers/pdf/SFB649DP2014-043.pdf Download aus dem Internet, Stand: 15.09.2014, Volltext: http://hdl.handle.net/10419/103778 
|
| Author Notes: | Enno Mammen, Christoph Rothe, and Melanie Schienle |
| Summary: | We study a general class of semiparametric estimators when the in nite-dimensional nuisance parameters include a conditional expectation function that has been estimated nonparametrically using generated covariates. Such estimators are used frequently to e.g. estimate nonlinear models with endogenous covariates when identification is achieved using control variable techniques. We study the asymptotic properties of estimators in this class, which is a non-standard problem due to the presence of generated covariates. We give conditions under which estimators are root-n consistent and asymptotically normal, derive a general formula for the asymptotic variance, and show how to establish validity of the bootstrap. |
|---|---|
| Physical Description: | Online Resource |
| Format: | Systemvoraussetzungen: Acrobat Reader. |