We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

econ.TH

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Economics > Theoretical Economics

Title: Correlated Choice

Abstract: As a means for testing whether a group of agents jointly maximize random utility, we introduce the correlated random utility model. The correlated random utility model asks that agents face correlated random draws of preferences which govern their decisions. We study joint random utility maximization through the lens of joint stochastic choice data (correlated choice rule), a novel type of data to the stochastic choice framework. Key is the property of marginality, which demands the independence of any given agent's marginal choices from the budgets faced by the remaining agents. Marginality permits the construction of well-defined marginal stochastic choice functions. Marginality and non-negativity of an analogue of the Block-Marschak polynomials characterize joint random utility maximization for small environments. For larger environments, we offer an example of a correlated choice rule establishing that each of the marginal stochastic choice rule may be stochastically rational while the correlated choice rule is not.
Comments: 53 pages, 2 figures
Subjects: Theoretical Economics (econ.TH)
Cite as: arXiv:2103.05084 [econ.TH]
  (or arXiv:2103.05084v2 [econ.TH] for this version)

Submission history

From: Christopher Turansick [view email]
[v1] Mon, 8 Mar 2021 21:28:18 GMT (34kb)
[v2] Tue, 10 Aug 2021 20:06:39 GMT (36kb)

Link back to: arXiv, form interface, contact.