Kybernetika 47 no. 1, 15-37, 2011

On the compound Poisson-gamma distribution

Christopher Withers and Saralees Nadarajah

Abstract:

The compound Poisson-gamma variable is the sum of a random sample from a gamma distribution with sample size an independent Poisson random variable. It has received wide ranging applications. In this note, we give an account of its mathematical properties including estimation procedures by the methods of moments and maximum likelihood. Most of the properties given are hitherto unknown.

Keywords:

compound Poisson-gamma, estimation, expansions, moments

Classification:

62E15, 62E17, 62E20

paper.pdf

References:

  1. T. A. Buishand: Stochastic Modelling of Daily Rainfall Sequences. Wageningen, Netherlands, Mededelingen Landbouwhogeschool 1977.   CrossRef
  2. L. Choo and S. G. Walker: A new approach to investigating spatial variations of disease. J. Roy. Statist. Soc. A 171 (2008), 395-405.   CrossRef
  3. A. Christensen, H. Melgaard and J. Iwersen: Environmental monitoring based on a hierarchical Poisson-gamma model. J. Quality Technology 35 (2003), 275-285.   CrossRef
  4. L. Comtet: Advanced Combinatorics. Reidel Publishing Company, Dordrecht 1974.   CrossRef
  5. R. A. Fisher and E. A. Cornish: The percentile points of distributions having known cumulants. Technometrics 2 (1960), 209-225.   CrossRef
  6. T. Fukasawa and I. V. Basawa: Estimation for a class of generalized state-space time series models. Statist. Probab. Lett. 60 (2002), 459-473.   CrossRef
  7. L. Galue: A generalized hyper Poisson-gamma distribution associated with the $H$-function. Hadronic J. 30 (2007), 63-79.   CrossRef
  8. I. S. Gradshteyn and I. M. Ryzhik: Tables of Integrals, Series and Products. Fourth edition. Academic Press, New York 1965.   CrossRef
  9. P. Hadjicostas and S. M. Berry: Improper and proper posteriors with improper priors in a Poisson-gamma hierarchical model. Test 8 (1999), 147-166.   CrossRef
  10. R. Henderson and S. Shimakura: A serially correlated gamma frailty model for longitudinal count data. Biometrika 90 (2003), 355-366.   CrossRef
  11. M. Kendall and A. Stuart: The Advanced Theory of Statistics. Volume 1. MacMillan, New York 1977.   CrossRef
  12. L. Le Cam: A stochastic description of precipitation. In: Proc. Fourth Berkeley Symposium on Mathematical Statistics and Probability (J. Neyman, ed.), University of California Press, Berkeley 1961, volume 3, pp. 165-186.   CrossRef
  13. S. Nahmias and W. S. Demmy: The logarithmic Poisson gamma-distribution - a model for leadtime demand. Naval Research Logistics 29 (1982), 667-677.   CrossRef
  14. A. Ozturk: On the study of a probability distribution for precipitation totals. J. Appl. Meteorology 20 (1981), 1499-1505.   CrossRef
  15. K. J. A. Revfeim: Comments ``On the study of a probability distribution for precipitation totals''. J. Appl. Meteology 21 (1982), 97-100.   CrossRef
  16. K. J. A. Revfeim: A theoretically derived distribution for annual rainfall totals. Internat. J. Climatology 10 (1990), 647-650.   CrossRef
  17. C. S. Withers: Asymptotic expansions for distributions and quantiles with power series cumulants. J. Roy. Statist. Soc. B 46 (1984), 389-396.   CrossRef
  18. C. S. Withers and S. Nadarajah: Saddlepoint Expansions in Terms of Bell Polynomials. Technical Report, Applied Mathematics Group, Industrial Research Ltd., Lower Hutt, New Zealand 2010. Avaiable on-line at http://arxiv.org/   CrossRef
  19. N. Xia, Z.-Z. Zhang and Z.-L. Ying: Convergence rate of the L-N estimator in Poisson-gamma models. Acta Math. Appl. Sinica 22 (2006), 639-654.   CrossRef