This paper describes our initial reactions to the inverse Gaussian-Poisson distribution (IGPD) recently introduced into bibliometrics by Sichel. We find that Sichel's claim that the IGPD fits many bibliometric data sets to be fully justified though our difficulties in applying it are described. A problem arising from our attempts to find its transform into a ranked distribution is also briefly discussed.
Get full access to this article
View all access options for this article.
References
1.
H.S. Sichel , On a distribution law for word frequencies , J. Amer. Statist. Assoc.70 (351) (1975) 542-547.
2.
H.S. Sichel , A bibliometric distribution which really works, J. Amer. Soc. Information Science36 (5) (1985) 314-321.
3.
H.S. Sichel , The GIGP distribution models with applications to physics literature, Czech. J. Physics B 36 (1) ( 1986) 133-137.
4.
M.C.K. Tweedie , Statistical properties of Inverse Gaussian distributions . I, Annals of Mathematical Statistics28 (1957) 362-373.
5.
P.A.P. Moran , An Introduction to Probability Theory (Oxford, 1968).
6.
I.J. Good, The population frequencies of species and the estimation of population parameters , Biometrika40 (3, 4) (1953) 237-264.
7.
B.C. Brookes , Frequency rank vs. frequency distributions: an information loss, Czech. J. Physics B 36 (1986) 30-32.
8.
S.D. Haitun , Stationary scientometrics distributions, Scientometric4 (1982) Part I: 5-25; Part II: 89-104; Part III: 181-194.
9.
B.C. Brookes , Research Note, Journal of Information Science.10 (2) (1985) 91-92, and 10 (3) (1985) 139.
