Department of Economics and Business Economics

Higher-order properties of approximate estimators

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Higher-order properties of approximate estimators. / Kristensen, Dennis; Salanié, Bernard.

In: Journal of Econometrics, Vol. 198, No. 2, 01.06.2017, p. 189-208.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

Kristensen, D & Salanié, B 2017, 'Higher-order properties of approximate estimators', Journal of Econometrics, vol. 198, no. 2, pp. 189-208. https://doi.org/10.1016/j.jeconom.2016.10.008

APA

CBE

MLA

Vancouver

Kristensen D, Salanié B. Higher-order properties of approximate estimators. Journal of Econometrics. 2017 Jun 1;198(2):189-208. doi: 10.1016/j.jeconom.2016.10.008

Author

Kristensen, Dennis ; Salanié, Bernard. / Higher-order properties of approximate estimators. In: Journal of Econometrics. 2017 ; Vol. 198, No. 2. pp. 189-208.

Bibtex

@article{0a5d25ed419145df8b1990f619a45a8b,
title = "Higher-order properties of approximate estimators",
abstract = "Many modern estimation methods in econometrics approximate an objective function, for instance, through simulation or discretization. These approximations typically affect both bias and variance of the resulting estimator. We first provide a higher-order expansion of such “approximate” estimators that takes into account the errors due to the use of approximations. We show how a Newton–Raphson adjustment can reduce the impact of approximations. Then we use our expansions to develop inferential tools that take into account approximation errors: we propose adjustments of the approximate estimator that remove its first-order bias and adjust its standard errors. These corrections apply to a class of approximate estimators that includes all known simulation-based procedures. A Monte Carlo simulation on the mixed logit model shows that our proposed adjustments can yield significant improvements at a low computational cost.",
keywords = "Bias adjustment, Extremum estimators, Higher-order expansion, Numerical approximation, Simulation-based estimation",
author = "Dennis Kristensen and Bernard Salani{\'e}",
year = "2017",
month = jun,
day = "1",
doi = "10.1016/j.jeconom.2016.10.008",
language = "English",
volume = "198",
pages = "189--208",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "2",

}

RIS

TY - JOUR

T1 - Higher-order properties of approximate estimators

AU - Kristensen, Dennis

AU - Salanié, Bernard

PY - 2017/6/1

Y1 - 2017/6/1

N2 - Many modern estimation methods in econometrics approximate an objective function, for instance, through simulation or discretization. These approximations typically affect both bias and variance of the resulting estimator. We first provide a higher-order expansion of such “approximate” estimators that takes into account the errors due to the use of approximations. We show how a Newton–Raphson adjustment can reduce the impact of approximations. Then we use our expansions to develop inferential tools that take into account approximation errors: we propose adjustments of the approximate estimator that remove its first-order bias and adjust its standard errors. These corrections apply to a class of approximate estimators that includes all known simulation-based procedures. A Monte Carlo simulation on the mixed logit model shows that our proposed adjustments can yield significant improvements at a low computational cost.

AB - Many modern estimation methods in econometrics approximate an objective function, for instance, through simulation or discretization. These approximations typically affect both bias and variance of the resulting estimator. We first provide a higher-order expansion of such “approximate” estimators that takes into account the errors due to the use of approximations. We show how a Newton–Raphson adjustment can reduce the impact of approximations. Then we use our expansions to develop inferential tools that take into account approximation errors: we propose adjustments of the approximate estimator that remove its first-order bias and adjust its standard errors. These corrections apply to a class of approximate estimators that includes all known simulation-based procedures. A Monte Carlo simulation on the mixed logit model shows that our proposed adjustments can yield significant improvements at a low computational cost.

KW - Bias adjustment

KW - Extremum estimators

KW - Higher-order expansion

KW - Numerical approximation

KW - Simulation-based estimation

U2 - 10.1016/j.jeconom.2016.10.008

DO - 10.1016/j.jeconom.2016.10.008

M3 - Journal article

AN - SCOPUS:85015363417

VL - 198

SP - 189

EP - 208

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -