Abstract
Results of Jain and Khan (1979) and Khan and Jain (1978) on the time to first emptiness of a reservoir are generalized to include the case of defective random variables where the mass at ∞ can be positive. The assumption of an underlying exponential family is not needed — the general condition is infinite divisibility and closure under convolutions. The support of the distributions can be nonnegative reals or nonnegative integers. Examples are given to illustrate the general theory and show the bivariate extension.
Similar content being viewed by others
References
Consul, P.C. (1989) Generalized Poisson distribution: Properties and applications. Marcel Dekker, New York.
Hassairi, A. (1999). Generalized variance and exponential families. Ann. Statist., 27, 374–385.
Jain, G.C. and Consul, P.C. (1971). A generalized negative binomial distribution. SIAM J. Appl. Math., 21, 501–513.
Jain, G.C. and Khan, M.S.H. (1979). On an exponential family. Math. Operationsforsch. Statist., Ser. Statist., 10, 153–168.
Johnson, N.L., Kotz, S. and Kemp, A.W. (1994). Univariate discrete distributions. Wiley, New York.
Kendall, D.G. (1957). Some problems in the theory of dams. J. Roy. Statist. Soc. Ser. B, 2, 207–212.
Khan, M.S.H. and Jain, G.C. (1978). A class of distributions in the first emptiness of a semi-infinite reservoir. Biometrical J. Zeitschrift, 20, 243–252.
Letac, G. and Mora, M. (1990). Natural exponential families with cubic variance functions. Ann. Statist., 18, 1–37.
Marshall, A.W. and Olkin, I. (2007). Life distributions: Structure of nonparametric, semiparametric, and parametric families. Springer, New York.
Morris, C.N. (1982). Natural exponential families with quadratic variance functions. Ann. Statist., 10, 65–80.
Seshadri, V. (1993). The inverse Gaussian distribution. A case study in exponential families. Clarendon, Oxford.
Seshadri, V. (1999). The inverse Gaussian distribution. Statistical theory and applications. Lecture Notes in Statistics, v. 137, Springer, New York.
Vandal, A.C. (1994). A compendium of variance functions for real natural exponential families. M.Sc. thesis, McGill University.
Acknowledgement
Thanks to Alain Vandal for providing his M.Sc. thesis.
This research was supported by an NSERC Discovery Grant.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 A.1 Natural exponential family; an example
Let f 0(z) be a density (say, with respect to Lebesgue or counting measure) with cumulant generating function ψ(t) over a range t ∈ Θ where the moment generating function e ψ is finite. Then, as in Morris (1982), the exponential family
can be derived. Its mean and variance are ψ′(θ) and ψ′′(θ) respectively. If f 0(z) = f 0(z;λ) is infinitely divisible with convolution parameter λ, then the cumulant generating function has the form of ψ(t) = ψ λ (t) = λψ 1(t), where ψ 1(t) is the cumulant generating function of f 0(z;1). The exponential family with “natural parameter” θ and convolution parameter λ > 0 is
For non-Poisson families, a non-convolution parameter appears in f 0 and it combines with θ in (A.1). If f 0(z;λ) is Poisson(λ), then (A.1) is Poisson(λe θ). From its definition, it follows that (A.1) has cumulant generating function λ[ψ 1(θ + t) − ψ 1(t)] for t such that θ ∈ Θ, θ + t ∈ Θ; furthermore its mean and variance are respectively λψ′1(θ) and λψ′′1(θ) for θ ∈ Θ.
This is illustrated for the negative binomial distribution NB(λ,ξ) with 0 < ξ < 1 fixed and q = 1 − ξ. The probability mass function is
and its cumulant generating function is \(\psi_\lambda(t)= \lambda[\log \xi-\log(1-qe^t)]\) and Θ = {t: qe t < 1}. The natural exponential family in (A.1) is
That is, this is the NB(λ,1 − qe θ) distribution and the natural exponential family generated from NB(λ,ξ) is NB with a different parametrization of the non-convolution parameter. For the parametrization of (A.2), the mean is μ(θ) = λqe θ/(1 − qe θ), the variance is V(θ) = λqe θ/(1 − qe θ)2 and V(θ) = μ(θ) + μ 2(θ)/λ for all θ ∈ Θ.
1.2 A.2 Results on derivatives of univariate and bivariate Laplace transforms
For the first two results, let Y be a nonnegative random variable with finite mean μ Y and LT L Y (s).
Result A1. |L′ Y (s)/L Y (s)| ≤ μ Y for all s ≥ 0.
Proof
Note that \(|L'_Y(s)|/L_Y(s)=\int_0^\infty ye^{-sy}dF_Y(y)/L_Y(s)\). Consider the family of densities with \(g(y;s)=e^{-sy}f_Y(y)/L_Y(s)\) on Lebesgue or counting measure. As s increases, the density g(·;s) puts less weight on large values of y. Hence g(·;s) is stochastic decreasing as s increases (from page 261 of Marshall and Olkin 2007, this family is decreasing in the stronger likelihood ratio order). Hence the expected value of Y * with density g(·;s) is decreasing in s. The expected value for s = 0 is μ Y and hence this is a bound.
Result B1. lim s→ ∞ |L′ Y (s)/L Y (s)| = 0.
Proof
For densities g(·;s) defined above, the family converges in distribution to the degenerate random variable 0 as s→ ∞.
The above results can be extended to bivariate LTs with similar proofs. Let (Y 1,Y 2) be a nonnegative random pair with finite means \(\mu_{Y_1},\mu_{Y_2}\) and LT \(L_{Y_1,Y_2}(s_1,s_2)\).
Result A2.
j = 1,2, for all s 1,s 2 ≥ 0.
Proof
Consider the family of densities with
on Lebesgue or counting measure. As s 1,s 2 increase, the density g(·;s 1,s 2) puts less weight on large values of y 1,y 2. Hence g(·;s 1,s 2) is stochastic decreasing as s 1,s 2 increase. Hence the expected values of random variables \(Y_1^*,Y_2^*\) with density g(·;s 1,s 2) are decreasing in s 1,s 2. The expected value for s 1 = s 2 = 0 are \(\mu_{Y_1},\mu_{Y_2}\) and hence these are bounds.
Result B2.
Proof
For densities g(·;s 1,s 2) defined above, the family converges in distribution to the degenerate random vector (0,0) as s→ ∞ with s 1 = sr 1 and s 2 = sr 2.
Rights and permissions
About this article
Cite this article
Joe, H., Seshadri, V. Infinitely divisible distributions arising from first crossing times and related results. Sankhya A 74, 222–248 (2012). https://doi.org/10.1007/s13171-012-0002-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13171-012-0002-z
Keywords and phrases
- Convolution-closed
- defective distribution
- generalized negative binomial
- generalized Poisson
- Kendall-Ressel distribution
- natural exponential family
- strict arcsine distribution