Semiparametric estimation with generated covariates

Arbeitspapier

We study a general class of semiparametric estimators when the infinite-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.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: KIT Working Paper Series in Economics ; No. 81

Classification and Topics

• Classification: Wirtschaft ; Semiparametric and Nonparametric Methods: General ; Multiple or Simultaneous Equation Models: Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions; Social Interaction Models
• Subject: semiparametric estimation ; generated covariates ; profiling ; propensity score

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Mammen, Enno ; Rothe, Christoph ; Schienle, Melanie
• Event: Veröffentlichung
• (who): Karlsruher Institut für Technologie (KIT), Institut für Volkswirtschaftslehre (ECON)
• (where): Karlsruhe
• (when): 2016

Further information

• DOI: doi:10.5445/IR/1000051816
• Handle: http://hdl.handle.net/10419/126096
• URN: urn:nbn:de:swb:90-518162
• Last update: 10.03.2025, 11:45 AM CET