Estimating the Link Function in Multinomial Response Models under Endogeneity and Quadratic Loss

George Judge () and Ron Mittelhammer ()

No 25095, CUDARE Working Papers from University of California, Berkeley, Department of Agricultural and Resource Economics

Abstract: This paper considers estimation and inference for the multinomial response model in the case where endogenous variables are arguments of the unknown link function. Semiparametric estimators are proposed that avoid the parametric assumptions underlying the likelihood approach as well as the loss of precision when using nonparametric estimation. A data based shrinkage estimator that seeks an optimal combination of estimators and results in superior risk performance under quadratic loss is also developed.

Keywords: Research; Methods/; Statistical; Methods (search for similar items in EconPapers)
Pages: 26
Date: 2004
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://ageconsearch.umn.edu/record/25095/files/wp040970.pdf (application/pdf)

Related works:
Working Paper: Estimating the Link Function in Multinomial Response Models under Endogeneity and Quadratic Loss (2004)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ags:ucbecw:25095

DOI: 10.22004/ag.econ.25095

Access Statistics for this paper

More papers in CUDARE Working Papers from University of California, Berkeley, Department of Agricultural and Resource Economics Contact information at EDIRC.
Bibliographic data for series maintained by AgEcon Search ().