The Lauricella theory of multiple hypergeometric functions is used to shed some light on certain distributional properties of the mean of a Dirichlet process. This approach leads to several results, which are illustrated here. Among these are a new and more direct procedure for determining the exact form of the distribution of the mean, a correspondence between the distribution of the mean and the parameter of a Dirichlet process, a characterization of the family of Cauchy distributions as the set of the fixed points of this correspondence, and an extension of the Markov–Krein identity. Moreover, an expression of the characteristic function of the mean of a Dirichlet process is obtained by resorting to an integral representation of a confluent form of the fourth Lauricella function. This expression is then employed to prove that the distribution of the mean of a Dirichlet process is symmetric if and only if the parameter of the process is symmetric, and to provide a new expression of the moment generating function of the variance of a Dirichlet process.
References
Akhiezer, N. I. and Krein, M. (1962). Some Questions in the Theory of Moments. Amer. Math. Soc., Providence, RI.
Mathematical Reviews (MathSciNet):
MR167806
Bondesson, L. (1992). Generalized Gamma Convolutions and Related Classes of Distributions and Densities. Springer, New York.
Carlson, B. C. (1977). Special Functions of Applied Mathematics. Academic Press, New York.
Mathematical Reviews (MathSciNet):
MR590943
Cifarelli, D. M. and Melilli, E. (2000). Some new results for Dirichlet priors. Ann. Statist. 28 1390--1413.
Cifarelli, D. M. and Regazzini, E. (1979a). Considerazioni generali sull'impostazione bayesiana di problemi non parametrici. Le medie associative nel contesto del processo aleatorio di Dirichlet. Parte $I$. Riv. Mat. Sci. Econom. Social. 2 39--52.
Mathematical Reviews (MathSciNet):
MR573686
Cifarelli, D. M. and Regazzini, E. (1979b). Considerazioni generali sull'impostazione bayesiana di problemi non parametrici. Le medie associative nel contesto del processo aleatorio di Dirichlet. Parte II. Riv. Mat. Sci. Econom. Social. 2 95--111.
Mathematical Reviews (MathSciNet):
MR574040
Cifarelli, D. M. and Regazzini, E. (1990). Distribution functions of means of a Dirichlet process. Ann. Statist. 18 429--442. [Correction (1994) Ann. Statist. 22 1633--1634.]
Cifarelli, D. M. and Regazzini, E. (1996). Tail-behaviour and finiteness of means of distributions chosen from a Dirichlet process. Technical Report IAMI 96.19, CNR-IAMI, Milano. Available at www.iami.mi.cnr.it/iami/reports.html.
Diaconis, P. and Kemperman, J. (1996). Some new tools on Dirichlet priors. In Bayesian Statistics 5 (J. M. Bernardo, J. O. Berger, A. P. Dawid and A. F. M. Smith, eds.) 97--106. Oxford Univ. Press.
Dickey, J. M. (1968). Three multidimensional-integral identities with Bayesian applications. Ann. Math. Statist. 39 1615--1628.
Mathematical Reviews (MathSciNet):
MR234545
Dickey, J. M. (1983). Multiple hypergeometric functions: Probabilistic interpretations and statistical uses. J. Amer. Statist. Assoc. 78 628--637.
Mathematical Reviews (MathSciNet):
MR721212
Dickey, J. M., Jiang, J. M. and Kadane, J. B. (1987). Bayesian methods for censored categorical data. J. Amer. Statist. Assoc. 82 773--781.
Mathematical Reviews (MathSciNet):
MR909982
Erdélyi, A. (1937). Beitrag zur theorie der konfluenten hypergeometrischen funktionen von mehreren veränderlichen. S. Akad. Wiss. Wien. 146 431--467.
Exton, H. (1976). Multiple Hypergeometric Functions and Applications. Wiley, New York.
Mathematical Reviews (MathSciNet):
MR422713
Feigin, P. B. and Tweedie, R. L. (1989). Linear functionals and Markov chains associated with Dirichlet processes. Math. Proc. Cambridge Philos. Soc. 105 579--585.
Mathematical Reviews (MathSciNet):
MR985694
Ferguson, T. S. (1973). A Bayesian analysis of some nonparametric problems. Ann. Statist. 1 209--230.
Mathematical Reviews (MathSciNet):
MR350949
Firmani, E. (2002). Distribuzione di probabilità esatta di un vettore di medie di un processo di Dirichlet e sua valutazione numerica. Degree thesis, Dept. Mathematics ``F. Casorati,'' Univ. Pavia.
Fristedt, B. and Gray, L. (1997). A Modern Approach to Probability Theory. Birkhäuser, Boston.
Henrici, P. (1991). Applied and Computational Complex Analysis 2. Wiley, New York.
Mathematical Reviews (MathSciNet):
MR372162
Hill, B. M. (1977). Exact and approximate Bayesian solutions for inference about variance components and multivariate inadmissibility. In New Developments in the Applications of Bayesian Methods (A. Aykac and C. Brumat, eds.) 129--152. North-Holland, Amsterdam.
Mathematical Reviews (MathSciNet):
MR505477
Jiang, J. M., Kadane, J. B. and Dickey, J. M. (1992). Computation of Carlson's multiple hypergeometric functions $\Rcr$ for Bayesian applications. J. Comput. Graph. Statist. 1 231--251.
Kerov, S. (1998). Interlacing measures. Amer. Math. Soc. Transl. Ser. 2 181 35--83.
Kerov, S. and Tsilevich, N. (1998). The Markov--Krein correspondence in several dimensions. PDMI Preprint 1/1998, Steklov Institute of Mathematics at St. Petersburg. Available at www.pdmi.ras.ru/~natalia/papers.html.
Lauricella, G. (1893). Sulle funzioni ipergeometriche a più variabili. Rend. Circ. Mat. Palermo 7 111--158.
Majumdar, S. (1992). On topological support of Dirichlet prior. Statist. Probab. Lett. 15 385--388.
Prudnikov, A. P., Brychkov, Yu. A. and Marichev, O. I. (1986). Integrals and Series 1. Elementary Functions. Gordon and Breach, Amsterdam.
Mathematical Reviews (MathSciNet):
MR874986
Regazzini, E. (1998). An example of the interplay between statistics and special functions. In Tricomi's Ideas and Contemporary Applied Mathematics 303--320. Accademia Nazionale dei Lincei, Rome.
Regazzini, E., Guglielmi, A. and Di Nunno, G. (2002). Theory and numerical analysis for exact distributions of functionals of a Dirichlet process. Ann. Statist. 30 1376--1411.
Slater, L. J. (1960). Confluent Hypergeometric Functions. Cambridge Univ. Press.
Mathematical Reviews (MathSciNet):
MR107026
Thorin, O. (1977a). On the infinite divisibility of the Pareto distribution. Scand. Actuar. J. 31--40.
Mathematical Reviews (MathSciNet):
MR431333
Thorin, O. (1977b). On the infinite divisibility of the lognormal distribution. Scand. Actuar. J. 121--148.
Mathematical Reviews (MathSciNet):
MR552135
Thorin, O. (1978a). An extension of the notion of a generalized $\Gamma$-convolution. Scand. Actuar. J. 141--149.
Mathematical Reviews (MathSciNet):
MR514310
Thorin, O. (1978b). Proof of a conjecture of L. Bondesson concerning the infinite divisibility of powers of a gamma variable. Scand. Actuar. J. 151--164.
Mathematical Reviews (MathSciNet):
MR514311
Tsilevich, N., Vershik, A. and Yor, M. (2000). Distinguished properties of the gamma processes and related properties. Prépublication du Laboratoire de Probabilités et Modèles Aléatoires 575, Université Paris 6. Available at www.pdmi.ras.ru/~natalia/papers.html.
Tsilevich, N., Vershik, A. and Yor, M. (2001). Remarks on the Markov--Krein identity and quasi-invariance of the gamma process. Zap. Nauchn. Sem. S. Petersburg. Otdel. Mat. Inst. Steklov. (POMI) 283 21--36.
Tucker, H. G. (1962). Absolute continuity of infinitely divisible distributions. Pacific J. Math. 12 1125--1129.
Mathematical Reviews (MathSciNet):
MR146868
Yamato, H. (1984). Characteristic functions of means of distributions chosen from a Dirichlet process. Ann. Probab. 12 262--267.
Mathematical Reviews (MathSciNet):
MR723745