A duration model with unobserved heterogeneity as a mixture of Dirichlet processes

Jan Ondrich; Kislaya Prasad

doi:10.1016/S0165-1765(97)00053-0

A duration model with unobserved heterogeneity as a mixture of Dirichlet processes

Jan Ondrich, Kislaya Prasad

Department of Economics

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

Abstract

We present a semiparametric procedure for the estimation of duration models in the presence of unobservable heterogeneity. The heterogeneity distribution is modelled as a mixture of Dirichlet processes. This places weak restrictions on its functional form and yields the popular Weibull model with Gamma mixing as a limiting case.Programs to estimate the model, as well as to generate the data, may be obtained on the internet at http://garnet.acns.fsu.edu/~kprasad/.

Original language	English (US)
Pages (from-to)	19-25
Number of pages	7
Journal	Economics Letters
Volume	55
Issue number	1
DOIs	https://doi.org/10.1016/S0165-1765(97)00053-0
State	Published - Aug 15 1997

Keywords

C14
C41
Dirichlet process
Duration model
Semiparametric estimation
Unobserved heterogeneity

ASJC Scopus subject areas

Finance
Economics and Econometrics

Access to Document

10.1016/S0165-1765(97)00053-0

Cite this

@article{d6514912d16345f18070dbfc678db399,

title = "A duration model with unobserved heterogeneity as a mixture of Dirichlet processes",

abstract = "We present a semiparametric procedure for the estimation of duration models in the presence of unobservable heterogeneity. The heterogeneity distribution is modelled as a mixture of Dirichlet processes. This places weak restrictions on its functional form and yields the popular Weibull model with Gamma mixing as a limiting case.Programs to estimate the model, as well as to generate the data, may be obtained on the internet at http://garnet.acns.fsu.edu/~kprasad/.",

keywords = "C14, C41, Dirichlet process, Duration model, Semiparametric estimation, Unobserved heterogeneity",

author = "Jan Ondrich and Kislaya Prasad",

year = "1997",

month = aug,

day = "15",

doi = "10.1016/S0165-1765(97)00053-0",

language = "English (US)",

volume = "55",

pages = "19--25",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - A duration model with unobserved heterogeneity as a mixture of Dirichlet processes

AU - Ondrich, Jan

AU - Prasad, Kislaya

PY - 1997/8/15

Y1 - 1997/8/15

N2 - We present a semiparametric procedure for the estimation of duration models in the presence of unobservable heterogeneity. The heterogeneity distribution is modelled as a mixture of Dirichlet processes. This places weak restrictions on its functional form and yields the popular Weibull model with Gamma mixing as a limiting case.Programs to estimate the model, as well as to generate the data, may be obtained on the internet at http://garnet.acns.fsu.edu/~kprasad/.

AB - We present a semiparametric procedure for the estimation of duration models in the presence of unobservable heterogeneity. The heterogeneity distribution is modelled as a mixture of Dirichlet processes. This places weak restrictions on its functional form and yields the popular Weibull model with Gamma mixing as a limiting case.Programs to estimate the model, as well as to generate the data, may be obtained on the internet at http://garnet.acns.fsu.edu/~kprasad/.

KW - C14

KW - C41

KW - Dirichlet process

KW - Duration model

KW - Semiparametric estimation

KW - Unobserved heterogeneity

UR - http://www.scopus.com/inward/record.url?scp=0031285235&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031285235&partnerID=8YFLogxK

U2 - 10.1016/S0165-1765(97)00053-0

DO - 10.1016/S0165-1765(97)00053-0

M3 - Article

AN - SCOPUS:0031285235

SN - 0165-1765

VL - 55

SP - 19

EP - 25

JO - Economics Letters

JF - Economics Letters

IS - 1

ER -

A duration model with unobserved heterogeneity as a mixture of Dirichlet processes

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this