Biometrika Advance Access originally published online on October 25, 2009
Biometrika 2009 96(4):761-780; doi:10.1093/biomet/asp053
Article |
Sinh-arcsinh distributions
Department of Mathematics and Statistics, The Open University, Walton Hall, Milton Keynes MK7 6AA, U.K. m.c.jones{at}open.ac.uk
Department of Mathematics, Escuela Politécnica, University of Extremadura, Avenida de la Universidad s/n, 10071 Cáceres, Spain apewsey{at}unex.es
Received for publication 1 August 2008. Revision received 1 April 2009.
We introduce the sinh-arcsinh transformation and hence, by applying it to a generating distribution with no parameters other than location and scale, usually the normal, a new family of sinh-arcsinh distributions. This four-parameter family has symmetric and skewed members and allows for tailweights that are both heavier and lighter than those of the generating distribution. The central place of the normal distribution in this family affords likelihood ratio tests of normality that are superior to the state-of-the-art in normality testing because of the range of alternatives against which they are very powerful. Likelihood ratio tests of symmetry are also available and are very successful. Three-parameter symmetric and asymmetric subfamilies of the full family are also of interest. Heavy-tailed symmetric sinh-arcsinh distributions behave like Johnson SU distributions, while their light-tailed counterparts behave like sinh-normal distributions, the sinh-arcsinh family allowing a seamless transition between the two, via the normal, controlled by a single parameter. The sinh-arcsinh family is very tractable and many properties are explored. Likelihood inference is pursued, including an attractive reparameterization. Illustrative examples are given. A multivariate version is considered. Options and extensions are discussed.
Key Words: Heavy tail • Johnson’s SUdistribution • Light tail • Sinh-normal distribution • Skew-normal distribution • Skewness • Transformation
References
-
Anderson T. W., Darling D. A. Asymptotic theory of certain goodness of fit criteria based on stochastic processes. Ann. Math. Statist. (1952) 23:193–212.[CrossRef]
Azzalini A. A class of distributions which includes the normal ones. Scand. J. Statist. (1985) 12:171–8.
Azzalini A. Further results on a class of distributions which includes the normal ones. Statistica (1986) 46:199–208.
Azzalini A., Capitanio A. Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t distribution. J. R. Statist. Soc. B (2003) 65:367–89.[CrossRef]
Azzalini A., Genton M. G. Robust likelihood methods based on the skew-t and related distributions. Int. Statist. Rev. (2008) 76:106–29.[CrossRef]
Barndorff-Nielsen O. E. Hyperbolic distributions and distributions on hyperbolae. Scand. J. Statist. (1978) 5:151–7.
Boos D. D. A test for asymmetry associated with the Hodges–Lehmann estimator. J. Am. Statist. Assoc. (1982) 77:647–51.[CrossRef][Web of Science]
Bowman K. O., Shenton L. R. Omnibus test contours for departures from normality based on 
and b2. Biometrika (1975) 62:243–50.
Box G. E. P., Tiao G. C. Bayesian Inference in Statistical Analysis (1973) Reading, MA: Addison-Wesley.
Branco M. D., Dey D. K. A general class of multivariate skew-elliptical distributions. J. Mult. Anal. (2001) 79:99–113.[CrossRef]
Cabilio P., Masaro J. A simple test of symmetry about an unknown median. Can. J. Statist. (1996) 24:349–61.[CrossRef]
Cassart D., Hallin M., Paindaveine D. Optimal detection of Fechner-asymmetry. J. Statist. Plan. Infer. (2008) 138:2499–525.[CrossRef]
D’Agostino R. B. An omnibus test of normality for moderate and large sample size. Biometrika (1971) 58:341–8.
D’Agostino R. B. Small sample probability points for the D test of normality. Biometrika (1972) 59:219–21.
DiCiccio T. J., Monti A. C. Inferential aspects of the skew exponential power distribution. J. Am. Statist. Assoc. (2004) 99:439–50.[CrossRef][Web of Science]
Downton F. Linear estimates with polynomial coefficients. Biometrika (1966) 53:129–41.
Fechner G. T. Kollectivmasslehre (1897) Leipzig: Engleman.
Fernández C., Steel M. F. J. On Bayesian modeling of fat tails and skewness. J. Am. Statist. Assoc. (1998) 93:359–71.[CrossRef][Web of Science]
Gebelein H. Das statistische Problem der Korrelation als Variations und Eigenwertproblem und sein Zusammenhang mit der Ausgleichungsrechnung. Z. Angew. Math. Mech. (1941) 21:364–79.[CrossRef]
Genton M. G. Skew-Elliptical Distributions and Their Applications: A Journey Beyond Normality (2004) Boca Raton, FL: Chapman and Hall/CRC.
Gradshteyn I. S., Ryzhik I. M. Table of Integrals, Series, and Products (1994) 5th ed. San Diego, CA: Academic Press.
Hoaglin D. C. Summarizing shape numerically: the g-and-h distributions. In: Exploring Data Tables, Trends and Shapes—Hoaglin D. C., Mosteller F., Tukey J. W., eds. (1986) New York: Wiley. 461–513.
Holland P. W., Wang Y. J. Dependence function for continuous bivariate densities. Commun. Statist. A (1987) 16:863–76.[CrossRef]
Jarque C., Bera A. Efficient tests for normality, homoscedasticity and serial independence of regression residuals. Economet. Lett. (1980) 6:255–9.[CrossRef]
Joe H. Multivariate Models and Dependence Concepts (1997) London: Chapman and Hall.
Johnson N. L. Systems of frequency curves generated by methods of translation. Biometrika (1949) 36:149–76.
Johnson N. L., Kotz S., Balakrishnan N. Continuous Univariate Distributions (1994) 1, 2nd ed. New York: Wiley.
Jones M. C. The local dependence function. Biometrika (1996) 83:899–904.
Jones M. C. Families of distributions arising from distributions of order statistics (with discussion). Test (2004) 13:1–43.[CrossRef][Web of Science]
Jones M. C. Contribution to the discussion of ‘Generalized additive models for location, scale and shape’, by R. A. Rigby and D. M. Stasinopoulos. Appl. Statist. (2005) 54:546–7.
Jones M. C., Faddy M. J. A skew extension of the t distribution, with applications. J. R. Statist. Soc. B (2003) 65:159–74.[CrossRef]
Kappenman R. F. Detection of symmetry or lack of it and applications. Commun. Statist. (1988) A 17:4163–77.[CrossRef]
Koyak R. A. On measuring internal dependence in a set of random variables. Ann. Statist. (1987) 15:1215–28.[CrossRef]
Ma Y., Genton M. G. Flexible class of skew-symmetric distributions. Scand. J. Statist. (2004) 31:459–68.[CrossRef]
Mudholkar G. S., Hutson A. The epsilon-skew-normal distribution for analyzing near-normal data. J. Statist. Plan. Infer. (2000) 83:291–309.[CrossRef]
Nelder J. A., Mead R. A simplex-method for function minimization. Comp. J. (1965) 7:308–13.
Rieck J. R., Nedelman J. R. A log-linear model for the Birnbaum–Saunders distribution. Technometrics (1991) 33:51–60.[CrossRef][Web of Science]
Royston P. AS R94: a remark on algorithm AS 181: the W-test for normality. Appl. Statist. (1995) 44:547–51.[CrossRef]
Samorodnitsky G., Taqqu M. S. Stable Non-Gaussian Random Processes: Stochastic Models with Infinite Variance (1994) Boca Raton, FL: Chapman and Hall/CRC.
Shapiro S., Wilk M. An analysis of variance test for normality (complete samples). Biometrika (1965) 52:591–611.
Smith R. L., Naylor J. C. A comparison of maximum likelihood and Bayesian estimators for the three-parameter Weibull distribution. Appl. Statist. (1987) 36:358–69.[CrossRef]
Stasinopoulos D. M., Rigby R. A. Generalized additive models for location scale and shape (GAMLSS). J. Statist. Software (2007) 23(7).
Tadikamalla P. R. Random sampling from the exponential power distribution. J. Am. Statist. Assoc. (1980) 75:683–6.[CrossRef][Web of Science]
Thadewald T., Büning H. Jarque–Bera test and its competitors for testing normality: a power comparison. J. Appl. Statist. (2007) 34:87–105.[CrossRef]
Tiku M. L., Islam M. Q., Selcuk A. S. Non-normal regression II – symmetric distributions. Commun. Statist. A (2001) 30:1021–45.[CrossRef]
van Zwet W. R. Convex Transformations of Random Variables (1964) Amsterdam: Mathematisch Centrum.
Zhang J., Wu Y. Likelihood-ratio tests for normality. Comp. Statist. Data Anal. (2005) 49:709–21.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||