Reliable estimation of random coefficient logit demand models

Daniel Brunner, Florian Heiss, Andre Romahn and Constantin Weiser

No 267, DICE Discussion Papers from Heinrich Heine University Düsseldorf, Düsseldorf Institute for Competition Economics (DICE)

Abstract: The differentiated demand model of Berry, Levinsohn and Pakes (1995) is widely used in empirical economic research. Previous literature has demonstrated numerical instabilities of the corresponding GMM estimator that give a wide range of parameter estimates and economic implications depending on technical details such as the choice of optimization algorithm, starting values, and convergence criteria. We show that these instabilities are mainly driven by numerical approximation errors of the moment function which is not analytically available. With accurate approximation, the estimator is well-behaved. We also discuss approaches to mitigate the computational burden of accurate approximation and provide code for download.

Date: 2017
New Economics Papers: this item is included in nep-dcm and nep-ecm
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (15)

Downloads: (external link)
https://www.econstor.eu/bitstream/10419/168359/1/897150325.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:zbw:dicedp:267

Access Statistics for this paper

More papers in DICE Discussion Papers from Heinrich Heine University Düsseldorf, Düsseldorf Institute for Competition Economics (DICE) Contact information at EDIRC.
Bibliographic data for series maintained by ZBW - Leibniz Information Centre for Economics ().