Bayesian Analysis

Specification of Informative Prior Distributions for Multinomial Models Using Vine Copulas

Kevin James Wilson

Advance publication

This article is in its final form and can be cited using the date of online publication and the DOI.

Full-text: Open access

Abstract

We consider the specification of an informative prior distribution for the probabilities in a multinomial model. We utilise vine copulas: flexible multivariate distributions built using bivariate copulas stacked in a tree structure. We take advantage of a specific vine structure, called a D-vine, to separate the specification of the multivariate prior distribution into that of marginal distributions for the probabilities and parameter values for the bivariate copulas in the vine. We provide guidance on each of the choices to be made in the prior specification and each of the questions to ask the expert to specify the model parameters within the context of an engineering application. We then give full details of the approach for the general problem.

Article information

Source
Bayesian Anal. (2017), 18 pages.

Dates
First available in Project Euclid: 9 October 2017

Permanent link to this document
https://projecteuclid.org/euclid.ba/1507536526

Digital Object Identifier
doi:10.1214/17-BA1068

Keywords
prior specification vines subjective Bayes multinomial elicitation

Rights
Creative Commons Attribution 4.0 International License.

Citation

Wilson, Kevin James. Specification of Informative Prior Distributions for Multinomial Models Using Vine Copulas. Bayesian Anal., advance publication, 9 October 2017. doi:10.1214/17-BA1068. https://projecteuclid.org/euclid.ba/1507536526


Export citation

References

  • Aas, K., Czado, C., Frigessi, A., and Bakken, H. (2009). “Pair-copula constructions of multiple dependence.”Insurance: Mathematics and Economics, 44: 182–198.
  • Bedford, T. and Cooke, R. (2002). “Vines – a new graphical model for dependent random variables.”Annals of Statistics, 30: 1031–1068.
  • Bedford, T., Daneshkhah, A., and Wilson, K. (2016). “Approximate Uncertainty Modeling in Risk Analysis with Vine Copulas.”Risk Analysis, 36: 792–815.
  • Chaloner, K. and Duncan, G. (1987). “Some properties of the Dirichlet-multinomial distribution and its use in prior elicitation.”Communications in Statistics-Theory and Methods, 16: 511–523.
  • Cooke, R., Kurowicka, D., and Wilson, K. (2015). “Sampling, Conditionalizing, Counting, Merging, Searching Regular Vines.”Journal of Multivariate Analysis, 138: 4–18.
  • Czado, C. and Min, A. (2011).Handbook on Vines, chapter Bayesian Inference for D-vines: Estimation and Model Selection. World Scientific.
  • Dickey, J. (1983). “Multiple hypergeometric functions: probabilistic interpretations of statistical uses.”Journal of the American Statistical Association, 78: 1615–1627.
  • Dißmann, J., Brechmann, E., Czado, C., and Kurowicka, D. (2013). “Selecting and estimating regular vine copulae and application to financial returns.”Computational Statistics and Data Analysis, 59: 52–69.
  • Elfadaly, F. and Garthwaite, P. (2013). “Eliciting Dirichlet and Connor-Mosimann prior distributions for multinomial models.”Test, 22: 628–646.
  • Elfadaly, F. and Garthwaite, P. (2016). “Eliciting Dirichlet and Gaussian copula prior distributions for multinomial models.”Statistics and Computing, 27: 449–467.
  • Garthwaite, P., Kadane, J., and O’Hagan, A. (2005). “Statistical methods for eliciting probability distributions.”Journal of the American Statistical Association, 100: 680–700.
  • Gruber, L. and Czado, C. (2015). “Sequential Bayesian Model Selection of Regular Vine Copulas.”Bayesian Analysis, 10: 937–963.
  • Haff, I., Aas, K., and Frigessi, A. (2010). “On the simplified pair-copula construction – simply useful or too simplistic?”Journal of Multivariate Analysis, 101: 1296–1310.
  • Joe, H. (1996).Distributions with Fixed Marginals and Related Topics, chapter Families of m-variate distributions with given margins and m(m-1)/2 bivariate dependence parameters. Institute of Mathematical Statistics.
  • Kurowicka, D. and Cooke, R. (2007). “Sampling algorithms for generating joint uniform distributions using the vine-copula method.”Computational Statistics and Data Analysis, 51: 2889–2906.
  • Min, A. and Czado, C. (2010a). “Bayesian model selection for D-vine pair-copula constructions.”Canadian Journal of Statistics, 39: 239–258.
  • Min, A. and Czado, C. (2010b). “Bayesian model selection for multivariate copulas using pair-copula constructions.”Journal of Financial Econometrics, 8: 511–546.
  • Min, A. and Czado, C. (2011). “Bayesian model selection for D-vine pair-copula constructions.”The Canadian Journal of Statistics, 39: 239–258.
  • O’Hagan, A., Buck, C., Daneshkhah, A., Eiser, J., Garthwaite, P., Jenkinson, D., Oakley, J., and Rakow, T. (2006).Uncertain Judgements: eliciting expert probabilities. Chichester: John Wiley.
  • Quigley, J. and Walls, L. (2010). “Reconciling experts opinion concerning the value of testing using Bayesian networks: a bridge too far?” In5th International ASRANet Conference.
  • Schepsmeier, U. (2015). “Efficient information based goodness-of-fit tests for vine copula models with fixed margins.”Journal of Multivariate Analysis, 138: 34–52.
  • Stöber, J., Joe, H., and Czado, C. (2013). “Simplified Pair Copula Constructions – Limitations and Extensions.”Journal of Multivariate Analysis, 119: 101–118.
  • van Dorp, J. and Mazzuchi, T. (2003). “Parameter specification of the beta distribution and its Dirichlet extensions utilizing quantiles.”Beta Distribution and Its Applications, 29: 1–37.
  • Wilson, K., Quigley, J., Bedford, T., and Walls, L. (2013). “Bayes linear Bayes graphical models in the design of optimal test strategies.”International Journal of Performability Engineering, 9: 715–728.
  • Wilson, K. J. (2017). “Supplementary Material to: Specification of informative prior distributions for multinomial models using vine copulas.”Bayesian Analysis.
  • Zapata-Vázquez, E., O’Hagan, A., and Bastos, L. (2014). “Eliciting expert judgements about a set of proportions.”Journal of Applied Statistics, 41: 1919–1933.

Supplemental materials