Partial identification in binary response models with nonignorable nonresponses

Tadao Hoshino

doi:10.1016/j.econlet.2013.07.009

Partial identification in binary response models with nonignorable nonresponses

Tadao Hoshino^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

This study investigates the identification of parameters in semiparametric binary response models of the form y=1(x'β+v+ε>0) when there are nonignorable nonresponses. We propose an estimation procedure for the identified set, the set of parameters that are observationally indistinguishable from the true value β, based on the special regressor approach of Lewbel (2000). We show that the estimator for the identified set is consistent in the Hausdorff metric.

Original language	English
Pages (from-to)	74-78
Number of pages	5
Journal	Economics Letters
Volume	121
Issue number	1
DOIs	https://doi.org/10.1016/j.econlet.2013.07.009
Publication status	Published - 2013 Aug 28
Externally published	Yes

Keywords

Nonignorable nonresponses
Partial identification
Semiparametric binary response models
Special regressor approach

ASJC Scopus subject areas

Finance
Economics and Econometrics

Access to Document

10.1016/j.econlet.2013.07.009

Cite this

@article{b0bc7aba5c944c97ba568cbf2d321302,

title = "Partial identification in binary response models with nonignorable nonresponses",

abstract = "This study investigates the identification of parameters in semiparametric binary response models of the form y=1(x'β+v+ε>0) when there are nonignorable nonresponses. We propose an estimation procedure for the identified set, the set of parameters that are observationally indistinguishable from the true value β, based on the special regressor approach of Lewbel (2000). We show that the estimator for the identified set is consistent in the Hausdorff metric.",

keywords = "Nonignorable nonresponses, Partial identification, Semiparametric binary response models, Special regressor approach",

author = "Tadao Hoshino",

year = "2013",

month = aug,

day = "28",

doi = "10.1016/j.econlet.2013.07.009",

language = "English",

volume = "121",

pages = "74--78",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Partial identification in binary response models with nonignorable nonresponses

AU - Hoshino, Tadao

PY - 2013/8/28

Y1 - 2013/8/28

N2 - This study investigates the identification of parameters in semiparametric binary response models of the form y=1(x'β+v+ε>0) when there are nonignorable nonresponses. We propose an estimation procedure for the identified set, the set of parameters that are observationally indistinguishable from the true value β, based on the special regressor approach of Lewbel (2000). We show that the estimator for the identified set is consistent in the Hausdorff metric.

AB - This study investigates the identification of parameters in semiparametric binary response models of the form y=1(x'β+v+ε>0) when there are nonignorable nonresponses. We propose an estimation procedure for the identified set, the set of parameters that are observationally indistinguishable from the true value β, based on the special regressor approach of Lewbel (2000). We show that the estimator for the identified set is consistent in the Hausdorff metric.

KW - Nonignorable nonresponses

KW - Partial identification

KW - Semiparametric binary response models

KW - Special regressor approach

UR - http://www.scopus.com/inward/record.url?scp=84882799536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882799536&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2013.07.009

DO - 10.1016/j.econlet.2013.07.009

M3 - Article

AN - SCOPUS:84882799536

SN - 0165-1765

VL - 121

SP - 74

EP - 78

JO - Economics Letters

JF - Economics Letters

IS - 1

ER -

Partial identification in binary response models with nonignorable nonresponses

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this