How McFadden Met Rockafellar and Learnt to Do More With Less

Research output: Working paper › Research

Standard

How McFadden Met Rockafellar and Learnt to Do More With Less. / Sørensen, Jesper Riis-Vestergaard; Fosgerau, Mogens.

2020.

Research output: Working paper › Research

Harvard

Sørensen, JR-V & Fosgerau, M 2020 'How McFadden Met Rockafellar and Learnt to Do More With Less'. https://doi.org/10.2139/ssrn.3581570

APA

Sørensen, J. R-V., & Fosgerau, M. (2020). How McFadden Met Rockafellar and Learnt to Do More With Less. University of Copenhagen. Institute of Economics. Discussion Papers (Online) No. 20-01 https://doi.org/10.2139/ssrn.3581570

Vancouver

Sørensen JR-V, Fosgerau M. How McFadden Met Rockafellar and Learnt to Do More With Less. 2020 Apr 30. https://doi.org/10.2139/ssrn.3581570

Author

Sørensen, Jesper Riis-Vestergaard ; Fosgerau, Mogens. / How McFadden Met Rockafellar and Learnt to Do More With Less. 2020. (University of Copenhagen. Institute of Economics. Discussion Papers (Online); No. 20-01).

Bibtex

@techreport{e6ef47e9b1ab4cb38b330d73e915a01d,
title = "How McFadden Met Rockafellar and Learnt to Do More With Less",
abstract = "We study the additive random utility model of discrete choice under minimal assumptions. We place no restrictions on the joint distribution of random utility components or the functional form of systematic utility components. Exploiting the power of convex analysis, we are nevertheless able to generalize a range of important results without resorting to differential theory. We characterize demand with a generalized Williams-Daly-Zachary theorem. A similarly generalized version of Hotz-Miller inversion yields constructive partial identication of systematic utilities for any known jointdistribution of stochastic utility components. Estimators based on our partial identication result remain well defined in the presence of zeros in demand. We also provide conditions for point identication, which are not only sufficient, but also necessary.",
keywords = "Faculty of Social Sciences, Additive random utility model, Discrete choice, Convex duality, Demand inversion, Partial identification, C25, C6, D11",
author = "S{\o}rensen, {Jesper Riis-Vestergaard} and Mogens Fosgerau",
year = "2020",
month = apr,
day = "30",
doi = "10.2139/ssrn.3581570",
language = "English",
series = "University of Copenhagen. Institute of Economics. Discussion Papers (Online)",
number = "20-01",
type = "WorkingPaper",

}

RIS

TY - UNPB

T1 - How McFadden Met Rockafellar and Learnt to Do More With Less

AU - Sørensen, Jesper Riis-Vestergaard

AU - Fosgerau, Mogens

PY - 2020/4/30

Y1 - 2020/4/30

N2 - We study the additive random utility model of discrete choice under minimal assumptions. We place no restrictions on the joint distribution of random utility components or the functional form of systematic utility components. Exploiting the power of convex analysis, we are nevertheless able to generalize a range of important results without resorting to differential theory. We characterize demand with a generalized Williams-Daly-Zachary theorem. A similarly generalized version of Hotz-Miller inversion yields constructive partial identication of systematic utilities for any known jointdistribution of stochastic utility components. Estimators based on our partial identication result remain well defined in the presence of zeros in demand. We also provide conditions for point identication, which are not only sufficient, but also necessary.

AB - We study the additive random utility model of discrete choice under minimal assumptions. We place no restrictions on the joint distribution of random utility components or the functional form of systematic utility components. Exploiting the power of convex analysis, we are nevertheless able to generalize a range of important results without resorting to differential theory. We characterize demand with a generalized Williams-Daly-Zachary theorem. A similarly generalized version of Hotz-Miller inversion yields constructive partial identication of systematic utilities for any known jointdistribution of stochastic utility components. Estimators based on our partial identication result remain well defined in the presence of zeros in demand. We also provide conditions for point identication, which are not only sufficient, but also necessary.

KW - Faculty of Social Sciences

KW - Additive random utility model

KW - Discrete choice

KW - Convex duality

KW - Demand inversion

KW - Partial identification

KW - C25

KW - C6

KW - D11

UR - https://www.mendeley.com/catalogue/7c4a7c18-a4f7-36c3-9896-d55df16ad08d/

U2 - 10.2139/ssrn.3581570

DO - 10.2139/ssrn.3581570

M3 - Working paper

T3 - University of Copenhagen. Institute of Economics. Discussion Papers (Online)

BT - How McFadden Met Rockafellar and Learnt to Do More With Less

ER -

ID: 248295261