Simulation-based estimation of peer effects

Brian Krauth

Econometrics from University Library of Munich, Germany

Abstract: The influence of peer behavior on an individual's choices has received renewed interest in recent years. However, accurate measures of this influence are difficult to obtain. Standard reduced-form methods lead to upwardly biased estimates due to simultaneity, common shocks, and nonrandom peer group selection. This paper describes a structural econometric model of peer effects in binary choice, as well as a simulated maximum likelihood estimator for its parameters. The model is nonparametrically identified under plausible restrictions, and can place informative bounds on parameter values under much weaker restrictions. Monte Carlo results indicate that this estimator performs better than a reduced form approach in a wide variety of settings. A brief application to youth smoking demonstrates the method and suggests that previous studies dramatically overstate peer influence.

Keywords: social interactions; peer effects; neighborhood effects (search for similar items in EconPapers)
JEL-codes: C1 C2 C3 C4 C5 C8 (search for similar items in EconPapers)
Pages: 40 pages
Date: 2004-08-02
New Economics Papers: this item is included in nep-dcm, nep-ecm and nep-ure
Note: Type of Document - pdf; pages: 40
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (30)

Downloads: (external link)
https://econwpa.ub.uni-muenchen.de/econ-wp/em/papers/0408/0408002.pdf (application/pdf)

Related works:
Journal Article: Simulation-based estimation of peer effects (2006)
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:wpa:wuwpem:0408002

Access Statistics for this paper

More papers in Econometrics from University Library of Munich, Germany
Bibliographic data for series maintained by EconWPA ( this e-mail address is bad, please contact ).