Obtaining analytic derivatives for a popular discrete-choice dynamic programming model

Curtis Eberwein; John C. Ham

doi:10.1016/j.econlet.2008.07.013

Obtaining analytic derivatives for a popular discrete-choice dynamic programming model

Curtis Eberwein, John C. Ham

Economics

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

Abstract

We show how to recursively calculate analytic first and second derivatives of the likelihood for a popular discrete-choice, dynamic programming model. These allow for decreased computing time, and are useful for de-bugging complicated program code and accurately estimating standard errors.

Original language	English (US)
Pages (from-to)	168-171
Number of pages	4
Journal	Economics Letters
Volume	101
Issue number	3
DOIs	https://doi.org/10.1016/j.econlet.2008.07.013
State	Published - Dec 2008

Keywords

Computation
Derivatives
Dynamic programming
Standard errors
Structural estimation

ASJC Scopus subject areas

Finance
Economics and Econometrics

Access to Document

10.1016/j.econlet.2008.07.013

Cite this

@article{b180d88297104825a6f63ccd2b3e92d6,

title = "Obtaining analytic derivatives for a popular discrete-choice dynamic programming model",

abstract = "We show how to recursively calculate analytic first and second derivatives of the likelihood for a popular discrete-choice, dynamic programming model. These allow for decreased computing time, and are useful for de-bugging complicated program code and accurately estimating standard errors.",

keywords = "Computation, Derivatives, Dynamic programming, Standard errors, Structural estimation",

author = "Curtis Eberwein and Ham, {John C.}",

note = "Funding Information: John Ham would like to thank the NSF for financial support. The authors would like to thank Donghoon Lee, Holger Sieg and Kenneth Wolpin for helpful comments. All mistakes are ours and the opinions expressed here in no way reflect those of the NSF. ",

year = "2008",

month = dec,

doi = "10.1016/j.econlet.2008.07.013",

language = "English (US)",

volume = "101",

pages = "168--171",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier B.V.",

number = "3",

}

TY - JOUR

T1 - Obtaining analytic derivatives for a popular discrete-choice dynamic programming model

AU - Eberwein, Curtis

AU - Ham, John C.

N1 - Funding Information: John Ham would like to thank the NSF for financial support. The authors would like to thank Donghoon Lee, Holger Sieg and Kenneth Wolpin for helpful comments. All mistakes are ours and the opinions expressed here in no way reflect those of the NSF.

PY - 2008/12

Y1 - 2008/12

N2 - We show how to recursively calculate analytic first and second derivatives of the likelihood for a popular discrete-choice, dynamic programming model. These allow for decreased computing time, and are useful for de-bugging complicated program code and accurately estimating standard errors.

AB - We show how to recursively calculate analytic first and second derivatives of the likelihood for a popular discrete-choice, dynamic programming model. These allow for decreased computing time, and are useful for de-bugging complicated program code and accurately estimating standard errors.

KW - Computation

KW - Derivatives

KW - Dynamic programming

KW - Standard errors

KW - Structural estimation

UR - http://www.scopus.com/inward/record.url?scp=55549097780&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=55549097780&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2008.07.013

DO - 10.1016/j.econlet.2008.07.013

M3 - Article

AN - SCOPUS:55549097780

SN - 0165-1765

VL - 101

SP - 168

EP - 171

JO - Economics Letters

JF - Economics Letters

IS - 3

ER -

Obtaining analytic derivatives for a popular discrete-choice dynamic programming model

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this