Annals of the Institute of Statistical Mathematics

, Volume 31, Issue 3, pp 373–390

# Generalized hypergeometric, digamma and trigamma distributions

Article

## Summary

In this paper we introduce and study new probability distributions named “digamma” and “trigamma” defined on the set of all positive integers. They are obtained as limits of the zero-truncated Type B3 generalized hypergeometric distributions (inverse Pólya-Eggenberger or negative binomial beta distributions), and also by compounding the logarithmic series distributions.

The family of digamma distributions has the logarithmic series as a limit and the trigamma as another limit. The trigamma distributions are very close to the zeta (Zipf) distributions. Thus, our new distributions are useful as substitutes of the logarithmic series when the observed frequency data have such a long tail that cannot be fitted by the latter distributions.

In the beginning sections we summarize properties of the Type B3 generalized hypergeometric distributions. It is emphasized that the distributions are obtained by compounding a Poisson distribution by “gamma product-ratio” distributions.

## Keywords

Hypergeometric Distribution Factorial Moment Logarithmic Series Entropy Distance Logarithmic Series Distribution

## References

1. [1]
Artin, E. (1964).The Gamma Function, Holt, Rinehart and Winston.Google Scholar
2. [2]
Boswell, M. T. and Patil, G. P. (1971). Chance mechanisms generating the logarithmic series distribution used in the analysis of number of species and number of individuals,Statistical Ecology, Vol. I, (eds. G. P. Patil, E. C. Pielu and W. E. Waters), Pennsylvania State University Press, 99–130.Google Scholar
3. [3]
de Medeiros, A. T. and Schwachheim, G. (1969). Polygamma functions with arbitrary precision, Algorithm 349,Commun. Ass. Comput. Math.,12, 213–214.Google Scholar
4. [4]
Irwin, J. O. (1965). Inverse factorial series as frequency distributions, Classical and Contagious Discrete DistributionsProc. Intnl. Symposium, Montreal, 1963, Calcutta Statistical Publ., 159–174.Google Scholar
5. [5]
Irwin, J. O. (1968). The generalized Waring distribution applied to accident theory,J. Roy. Statist. Soc., A,131, 205–225.
6. [6]
Irwin, J. O. (1975). The generalized Waring distribution,J. Roy. Statist. Soc., A,138, Part I, 18–31; Part II, 205–225; Part III, 374–384.
7. [7]
Johnson, N. L. and Kotz, S (1977).Urn Models and Their Application, John Wiley, New York.
8. [8]
Kendall, M. G. and Stuart, A. (1958).The Advanced Theory of Statistics, Vol. 1, Charles Griffin Co.Google Scholar
9. [9]
Markman, B. (1965). The Riemann zeta function,BIT,5, 138–141.Google Scholar
10. [10]
Matsunawa, T. (1976). Some inequalities based on inverse factorial series,Ann. Inst. Statist. Math.,28, 291–305.
11. [11]
Mizutani, S. (1953). Vocabulary in women's magazines,Kokken-Hokoku,4, National Language Research Institute, Tokyo, (in Japanese).Google Scholar
12. [12]
Nörlund, N. E. (1924).Vorlesungen über Differenzenrechnung, Springer, Berlin.
13. [13]
Ruben, H. (1976). A note on the tri-gamma function,Amer. Math. Monthly,83, 622–623.
14. [14]
Shimizu, R. (1968). Generalized hypergeometric distributions,Proc. Inst. Statist. Math.,16, 147–165, (in Japanese).
15. [15]
Sibuya, M. and Shimizu, R. (1979). Classification of generalized hypergeometric distributions, to appear.Google Scholar
16. [16]
Simon, H. A. (1955). On a class of skew distribution functions,Biometrika,42, 425–440.
17. [17]
Simon, H. A. (1960). Some further notes on a class of skew distribution functions,Inf. Control.,3, 80–88.
18. [18]
Slater, L. J. (1964). Confluent hypergeometric functionsHandbook of Mathematical Functions, (eds. Abramowitz, M. and Stegun, I. A.), Dover, New York.Google Scholar