Estimation issues for copulas applied to marketing data

Peter J. Danaher, Michael S. Smith

Research output: Contribution to journalArticleOtherpeer-review


Estimating copula models using Bayesian methods presents some subtle challenges, ranging from specification of the prior to computational tractability. There is also some debate about what is the most appropriate copula to employ from those available. We address these issues here and conclude by discussing further applications of copula models in marketing.

Original languageEnglish
Pages (from-to)25-28
Number of pages4
JournalMarketing Science
Issue number1
Publication statusPublished - Jan 2011
Externally publishedYes


  • Bayesian estimation
  • Discrete copula
  • Gaussian copula
  • Markov chain monte carlo
  • Media modeling
  • Probability models
  • Website page views

Cite this