Statistical Inference for Stochastic Processes

, Volume 8, Issue 3, pp 283–309 | Cite as

Bayesian Nonparametric Analysis for a Generalized Dirichlet Process Prior

  • Antonio LijoiEmail author
  • Ramsés H. Mena
  • Igor Prünster


This paper considers a generalization of the Dirichlet process which is obtained by suitably normalizing superposed independent gamma processes having increasing integer-valued scale parameter. A comprehensive treatment of this random probability measure is provided. We prove results concerning its finite-dimensional distributions, moments, predictive distributions and the distribution of its mean. Most expressions are given in terms of multiple hypergeometric functions, thus highlighting the interplay between Bayesian Nonparametrics and special functions. Finally, a suitable simulation algorithm is applied in order to compute quantities of statistical interest.


Bayesian nonparametric inference Dirichlet process generalized gamma convolutions Lauricella hypergeometric functions means of random probability measures predictive distributions 

AMS 200 Mathematics Subject Classifications

62F15 60G57 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Allaart, P. 2003Moments of the mean of Dubins-Freedman random probability distributionsJ. Theor. Probab.16471488CrossRefzbMATHMathSciNetGoogle Scholar
  2. Billingsley, P. 2003Probability and Measure3WileyNew YorkGoogle Scholar
  3. Bloomer, L., Hill, T.P. 2002Random probability measures with given mean and varianceJ. Theor. Probab.14919937MathSciNetGoogle Scholar
  4. Bondesson L. Generalized gamma convolutions and related classes of distributions and densities, Lecture Notes in Statistics Vol. 76, Springer, New York, 1992.Google Scholar
  5. Cifarelli, D. M. and Regazzini, E. Considerazioni generali sull’impostazione bayesiana di problemi non parametrici. Le medie associative nel contesto del processo aleatorio di Dirichlet Parte II, Rivista di matematica per le scienze economiche e sociali 1 (1979), 95–111.Google Scholar
  6. Cifarelli, D. M. and Regazzini, E. Distribution functions of means of a Dirichlet process, Ann. Stat. 18 (1990), 429–442 (Correction in Ann. Statist. 22 (1994), 1633–1634).Google Scholar
  7. Epifani, I., Lijoi, A., Prünster, I. 2003Exponential functionals and means of neutral-to-the-right priorsBiometrika90791808CrossRefMathSciNetGoogle Scholar
  8. Escobar, M.D. 1994Estimating normal means with a Dirichlet process priorJ. Am. Stat. Ass.89268277zbMATHMathSciNetGoogle Scholar
  9. Escobar, M.D., West, M. 1995Bayesian density estimation and inference using mixturesJ. Am. Stat. Ass.90577588MathSciNetGoogle Scholar
  10. Exton, H. 1976Multiple hypergeometric functions and applicationsEllis HorwoodChichesterGoogle Scholar
  11. Ferguson, T.S. 1973A Bayesian analysis of some nonparametric problemsAnn. Stat.1209230zbMATHMathSciNetGoogle Scholar
  12. Freedman, D.A. 1963On the asymptotic behavior of Bayes’ estimates in the discrete caseAnn. Math. Stat.3413861403Google Scholar
  13. Halmos, P. 1944Random almsAnn. Math. Stat.15182189zbMATHMathSciNetGoogle Scholar
  14. Hjort, N. L. Topics in non–parametric Bayesian statistics, In: P. J. Green et al. (eds), Highly Structured Stochastic Systems, Oxford University Press, Oxford, Oxford Statistical Science Series 27 (2003), 455–478.Google Scholar
  15. Ishwaran, H., James, L.F. 2001Gibbs sampling methods for stick-breaking priorsJ. Am. Stat. Ass.96161173MathSciNetGoogle Scholar
  16. Ishwaran, H., James, L.F. 2003Some further developments for stick–breaking priors: finite and infinite clustering and classificationSankhya65577592MathSciNetGoogle Scholar
  17. James, L. F. Poisson Process Partition Calculus with applications to exchangeable models and Bayesian Nonparametrics, Mathematics ArXiv math.PR/0205093 (2002).Google Scholar
  18. James, L. F., Lijoi, A. and Prünster, I. Conjugacy as a distinctive feature of the Dirichlet process, Scandinavian Journal of Statistics (2004), to appear.Google Scholar
  19. Kingman, J.F.C. 1975Random discrete distributionsJ. Roy Stat. Soc. Ser. B.37122zbMATHMathSciNetGoogle Scholar
  20. Lauricella, G. 1893Sulle funzioni ipergeometriche a piùvariabiliRend. Circ. Mat. Palermo7111158zbMATHGoogle Scholar
  21. Lijoi, A., Mena, R. H. and Prünster, I. Hierarchical mixture modelling with normalized inverse Gaussian priors, ICER Working Papers 12/04 (2004).Google Scholar
  22. Lijoi, A., Regazzini, E. 2004Means of a Dirichlet process and multiple hypergeometric functionsAnn. Probab.3214691495MathSciNetGoogle Scholar
  23. MacEachern, S.N. 1994Estimating normal means with a conjugate style Dirichlet process priorCommun. Statist. Simulation Comp.23727741zbMATHMathSciNetGoogle Scholar
  24. Muliere, P. and Secchi, P. A note on a proper Bayesian bootstrap, Quaderni del Dipartimento di Economia Politica e Metodi Quantitativi 18 (1995), Universtiàdegli Studi di Pavia.Google Scholar
  25. Nieto-Barajas, L. E., Prünster, I. and Walker, S. G. Normalized random measures driven by increasing additive processes, Ann Stat. 32 (2004), to appear.Google Scholar
  26. Petrone, S., Raftery, A.E. 1997A note on the Dirichlet process prior in Bayesian nonparametric inference with partial exchangeabilityStat. Prob. Lett.366983CrossRefMathSciNetGoogle Scholar
  27. Pitman, J. 1995Exchangeable and partially exchangeable random partitionsProb. Theory Related Fields102145158CrossRefzbMATHMathSciNetGoogle Scholar
  28. Pitman, J. Some developments of the Blackwell-MacQueen urn scheme, In: T. S. Ferguson et al. (eds), Statistics, probability and game theory. Papers in honor of David Blackwell, Institute of Mathematical Statistics, Hayward, Lecture Notes, Monograph Series Vol 30, 1996, pp. 245–267.Google Scholar
  29. Pitman, J. Poisson-Kingman partitions. In: Goldstein, D.R. (ed), Science and Statistics: A Festschrift for Terry Speed, Institute of Mathematical Statistics, Hayward, Lecture Notes, Monograph Series, Vol 40, 2003, pp. 1–35.Google Scholar
  30. Pitman, J., Yor, M. 1997The two-parameter Poisson-Dirichlet distribution derived from a stable subordinatorAnn Prob.25855900MathSciNetGoogle Scholar
  31. Prünster, I. Random probability measures derived from increasing additive processes and their application to Bayesian statistics, Ph.d dissertation, Universitàdegli Studi di Pavia, Pavia, 2002.Google Scholar
  32. Oldham, K.B., Spanier, J. 1974The fractional calculusAcademic PressNew YorkGoogle Scholar
  33. Regazzini, E. An example of the interplay between statistics and special functions, In: Tricomi’s ideas and contemporary applied mathematics, Acc. Naz. Lincei e Acc. Scienze Torino, Roma, Atti dei Convegni Lincei Vol. 147, 1998, pp. 303–320.Google Scholar
  34. Regazzini, E. Foundations of Bayesian statistics and some theory of Bayesian nonparametric methods, Lecture Notes, Stanford University, Stanford, 2001.Google Scholar
  35. Regazzini, E., Guglielmi, A., Di Nunno, G. 2002Theory and numerical analysis for exact distribution of functionals of a Dirichlet processAnn. Stat.3013761411MathSciNetGoogle Scholar
  36. Regazzini, E., Lijoi, A., Prünster, I. 2003Distributional results for means of random measures with independent incrementsAnn. Stat.31560585Google Scholar
  37. Sato, K. 1999Lévy processes and infinitely divisible distributionsCambridge University PressCambridgeGoogle Scholar
  38. Sethuraman, J. 1994A constructive definition of Dirichlet priorsStat. Sinica4639650zbMATHMathSciNetGoogle Scholar
  39. Skiena, S. 1990Implementing discrete mathematics: combinatorics and graph theory with MathematicaAddison-WesleyReading, MAGoogle Scholar
  40. Walker, S.G., Damien, P. 1998A full Bayesian non–parametric analysis involving a neutral to the right processScand. J Stat.25669680CrossRefMathSciNetGoogle Scholar
  41. Walker, S.G., Damien, P., Laud, P.W., Smith, A.F.M. 1999Bayesian nonparametric inference for random distributions and related functionsJ. Roy. Stat. Soc. Ser. B.61485527MathSciNetGoogle Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  • Antonio Lijoi
    • 1
    Email author
  • Ramsés H. Mena
    • 2
  • Igor Prünster
    • 3
    • 4
  1. 1.Dipartimento di Economia Politica e Metodi QuantitativiUniversità degli Studi di PaviaPaviaItaly
  2. 2.Departimento de Probabilidad y EstadisticaIIMAS-UNAM-MéxicoMexicoMexico
  3. 3.Dipartimento di Economia Politica e Metodi QuantitativiUniversità degli Studi di PaviaPaviaItaly
  4. 4.ICERTorinoItaly

Personalised recommendations