Loading [MathJax]/jax/output/SVG/fonts/TeX/fontdata.js
Variance
  • Menu
  • Articles
    • Articles in Press
    • Capital Management
    • Data Management and Information
    • Financial and Statistical Methods
    • Other
    • Ratemaking and Product Information
    • Reserving
    • Risk Management
    • All
  • For Authors
  • Editorial Board
  • About
  • Issues
  • Archives
  • search

    Sorry, something went wrong. Please try your search again.
    ×

    • Articles
    • Blog posts

RSS Feed

Enter the URL below into your favorite RSS reader.

https://variancejournal.org/feed
×
Financial and Statistical Methods
Vol. 16, Issue 1, 2023February 13, 2023 EDT

Discussion of “The Log-Gamma Distribution and Non-Normal Error”

Elias S. W. Shiu,
Log-gamma distribution cumulant T.N. Thiele linear exponential family
Photo by Markus Spiske on Unsplash
Variance
Shiu, Elias S. W. 2023. “Discussion of ‘The Log-Gamma Distribution and Non-Normal Error.’” Variance 16 (1).
Save article as...▾
  • PDF
  • XML
  • Citation (BibTeX)

View more stats

Abstract

A key goal of the discussion is to show a relatively simple way to derive the cumulant formulas in Appendix B of the paper “The Log-Gamma Distribution and Non-Normal Error.” It also suggests alternative derivation for various results in the paper. The concept of cumulants was introduced by T. N. Thiele. The discussion also points out some important actuarial roles played by Mr. Thiele.

There are many interesting historical facts in this paper. I would like to add the following. The concept of cumulants, which plays a prominent role in the paper, was introduced in 1899 by Thorvald Nicolai Thiele (1838-1910) under the name half-invariants, several decades before the rediscovery by the eminent statistician R. A. Fisher. Thiele was a Danish actuary, astronomer, mathematician, and statistician. He was a cofounder of the Danish insurance company Hafnia and was its Mathematical Director (actuary). Also, he was the founding president of the Danish Actuarial Society, a corresponding member of the British Institute of Actuaries, and a member of the Board of the Permanent Committee of the International Congresses of Actuaries. The net premium reserve differential equation,

ddttV=P+tVδ−(b−tV)μx+t,

in the theory of life contingencies was due to him, although it was not published until after he died. More information about Thiele can be found in the Encyclopedia of Actuarial Science article by Norberg (2004) or by looking up the website MacTutor History of Mathematics Archive (Robertson and O’Connor, n.d.).

Appendix B

Let me present a relatively simple way to derive the formulas in Appendix B. The cumulant-generating function of a random variable X is

ΨX(t):=ln[MX(t)].

Because κn(X), the nth cumulant of X, is the nth derivative of ΨX(t) at 0, κn(X) is the coefficient of tnn! in the Maclaurin series of ΨX(t). It turns out that it is not necessary to do “tedious” differentiations to determine the Maclaurin series because we have the logarithm series

ln(1+z)=z−z22+z33−z44±⋯

Observe that, with μ:=E[X],

MX(t)=E[etX]=etμE[et(X−μ)],

and that, with μj:=E[(X−μ)j],j=2,3,4,…,

E[et(X−μ)]=E[∑j≥0[t(X−μ)]jj!]=∑j≥0E[[t(X−μ)]jj!]=1+0+∑j≥2μjtjj!.

Hence, by using the logarithm series and some school algebra,

ΨX(t)=μt+ln(1+∑j≥2μjtjj!)=μt+∑j≥2μjtjj!−12(∑j≥2μjtjj!)2±⋯=μt+μ2t22!+μ3t33!+(μ4−3μ22)t44!+(μ5−10μ2μ3)t55!+⋯

Section 2

Let me derive the cumulant formulas on the right column of page 175 in a more general context. From the left column on page 175, one can see that the probability density function (pdf) of the log-gamma random variable Y has the form

fY(u)=eαu+g(u)+h(α),

which means that Y is a member of the linear exponential family, a concept that can be found in the actuarial textbook Klugman, Panjer, and Willmot (2019). Because a pdf integrates to 1, we have

∫∞−∞eαu+g(u)du=e−h(α).

Thus, the moment-generating function of Y is

MY(t)=∫∞−∞e(t+α)u+g(u)+h(α)du=e−h(t+α)+h(α),

which implies that the cumulant-generating function is

ΨY(t)=h(α)−h(t+α)=h(α)−∑j≥0h(j)(α)tjj!=−∑j≥1h(j)(α)tjj!.

Hence for the linear exponential family, we have for j=1,2,3,...,

κj=−h(j)(α);

also,

κj+1=∂∂ακj.

The actuarial textbook Kaas et al. (2008, 300) calls −h(α) the cumulant function. For the random variable Y in Section 2,

−h(α)=lnΓ(α)+αlnθ.

Section 4

There seems no need for the moment-generating function calculation in the right column of page 177. Here,

Y=γlnX1−γlnX2+C.

With X1 and X2 being independent random variables,

ΨY(t)=ΨγlnX1(t)+Ψ−γlnX2(t)+Ct=ΨlnX1(γt)+ΨlnX2(−γt)+Ct.

Because the cumulants of a random variable can be obtained from the coefficients of the Maclaurin series of the cumulant-generating function, it follows from the last equation that

κ1(Y)=γκ1(lnX1)−γκ1(lnX2)+C,κj(Y)=γjκj(lnX1)+(−γ)jκj(lnX2),j=2,3,4,….

In Section 4, X1 is assumed to be a Gamma(α,1) random variable, and X2 a Gamma(β,1) random variable. Hence,

κj(lnX1)=dj dαjlnΓ(α)

and

κj(lnX2)=dj dβjlnΓ(β).

Section 7

It is stated in the Conclusion paragraph of the paper that “loss distributions … are non-negative, positively skewed, and right-tailed.” Then it is claimed that “most loss distributions are transformations of the gamma distribution” which I find confusing. It may be useful for me to state the following result, whose proof can be found in a short discussion by Ko and Ng (2007). The distribution of any positive random variable can be arbitrarily closely approximated by a weighted average of Erlang distributions, where the weights are positive. An Erlang distribution is a gamma distribution with the shape parameter α being a positive integer.

Also, Ko and Ng (2007) show that the distribution of any positive random variable can be arbitrarily closely approximated by a weighted average of exponential distributions, but some of the weights may be negative. There are other approximation results given in this two-page discussion by Ko and Ng (2007).

Submitted: July 01, 2021 EDT

Accepted: October 04, 2021 EDT

References

Kaas, Rob, Marc Goovaerts, Jan Dhaene, and Michel Denuit. 2008. Modern Actuarial Risk Theory: Using R. 2nd ed. New York: Springer. https://doi.org/10.1007/978-3-540-70998-5.
Google Scholar
Klugman, S.A., H.H. Panjer, and G.E. Willmot. 2019. Loss Models: From Data to Decisions. 5th ed. New York: Wiley.
Google Scholar
Ko, B., and A.C.Y. Ng. 2007. “Discussion on ‘Stochastic Life Annuities’ by D. Dufresne.” North American Actuarial Journal 11 (3): 170–71.
Google Scholar
Norberg, R. 2004. “Thiele, T. N. (1838-1910).” In Encyclopedia of Actuarial Science, 1671–72. New York: Wiley.
Google Scholar
Robertson, E., and J. O’Connor. n.d. “MacTutor History of Mathematics Archive.” University of St. Andrews. https://mathshistory.st-andrews.ac.uk/.

Powered by Scholastica, the modern academic journal management system