Abstract
Four-parameter sinh–arcsinh classes provide flexible distributions with which to model skew, as well as light- or heavy-tailed, departures from a symmetric base distribution. A quantile-based method of estimating their parameters is proposed and the resulting estimates advocated as starting values from which to initiate maximum likelihood estimation. Parametric bootstrap edf-based goodness-of-fit tests for sinh–arcsinh distributions are proposed, and their operating characteristics for small- to medium-sized samples explored in Monte Carlo experiments. The developed methodology is illustrated in the analysis of data on the body mass index of athletes and the depth of snow on an Antarctic ice floe.
Similar content being viewed by others
References
Azzalini A (1985) A class of distributions which includes the normal ones. Scand J Stat 12:171–178
Azzalini A, Capitanio A (2014) The skew-normal and related families. Cambridge University Press, Cambridge
Babu GJ, Rao CR (2004) Goodness-of-fit tests when parameters are estimated. Sankhyā 66:63–74
Bagkeris E, Malyuta R, Volokha A, Cortina-Borja M, Bailey H, Townsend CL, Thorne C (2015) Pregnancy outcomes in HIV-positive women in Ukraine, 2000–12 (European Collaborative Study in EuroCoord): an observational cohort study. Lancet HIV 2:385–392
Balanda KP, MacGillivray HL (1988) Kurtosis: a critical review. Am Stat 42:111–119
Beran R (1986) Simulated power functions. Ann Stat 14:151–173
Cook RD, Weisberg S (1994) An introduction to regression graphics. Wiley, New York
D’Agostino RB, Stephens MA (eds) (1986) Goodness-of-fit techniques. Dekker, New York
Fischer M, Herrmann K (2013) The HS-SAS and GSH-SAS distribution as model for unconditional and conditional return distributions. Austrian J Stat 42:33–45
Fischer MJ (2014) Generalized hyperbolic secant distributions: with applications to finance. Springer, Heidelberg
Georgikopoulos NI, Voudouri V (2014) Demand dynamics and peer effects in consumption: historic evidence from a non-parametric model. Arch Econ Hist 26:27–59
Harkness WL, Harkness ML (1968) Generalized hyperbolic secant distributions. J Am Stat Assoc 63:329–337
Hope AC (1968) A simplified Monte Carlo significance test procedure. J R Stat Soc B 30:582–598
Jöckel K-H (1986) Finite sample properties and asymptotic efficiency of Monte Carlo tests. Ann Stat 14:336–347
Jones MC (2015) On families of distributions with shape parameters. Int Stat Rev 83:175–192
Jones MC, Pewsey A (2009) Sinh–arcsinh distributions. Biometrika 96:761–780
Jones MC, Rosco JF, Pewsey A (2011) Skewness-invariant measures of kurtosis. Am Stat 65:89–95
Ke X, Cortina-Borja M, Silva BC, Lowe R, Rakyan V, Dalding D (2013) Integrated analysis of genome-wide genetic and epigenetic association data for identification of disease mechanisms. Epigenetics 8:1236–1244
Knowles RL, Day T, Wade A, Bull C, Wren C, Dezateux C (2014) Patient-reported quality of life outcomes for children with serious congenital heart defects. Arch Dis Child 99:413–419
Marriott FH (1979) Barnard’s Monte Carlo tests: how many simulations? J R Stat Soc C 28:75–77
Matsumoto K, Voudouris V, Stasinopoulos D, Rigby R, Di Maio C (2012) Exploring crude oil production and export capacity of the OPEC Middle East countries. Energ Policy 48:820–828
McKeague IW (2015) Central limit theorems under special relativity. Stat Probab Lett 99:149–155
Nelder JA, Mead R (1965) A simplex method for function minimization. Comput J 7:308–313
Perks W (1932) On some experiments in the graduation of mortality statistics. J Inst Act 63:12–57
Pewsey A (2000) Problems of inference for Azzalini’s skew-normal distribution. J Appl Stat 27:859–870
Pewsey A, Abe T (2015) The sinh–arcsinhed logistic family of distributions: properties and inference. Ann Inst Stat Math 67:573–594
Pewsey A, Neuhäuser M, Ruxton GD (2013) Circular statistics in R. Oxford University Press, Oxford
Pingel R (2014) Some approximations of the logistic distribution with application to the covariance matrix of logistic regression. Stat Probab Lett 85:63–68
Romano JP (1988) A bootstrap revival of some nonparametric distance tests. J Am Stat Assoc 83:698–708
Rosco JF, Jones MC, Pewsey A (2011) Skew \(t\) distributions via the sinh–arcsinh transformation. Test 20:630–652
Rubio FJ, Ogundimu EO, Hutton JL (2016) On modelling asymmetric data using two-piece sinh–arcsinh distributions. Braz J Probab Stat 30:485–501
Santos-Fernández E, Govindaraju K, Jones G (2014) A new variables acceptance sampling plan for food safety. Food Control 44:249–257
Spinelli JJ, Stephens MA (1983) Tests for exponentiality when origin and scale parameters are unknown. Technometrics 29:471–476
Stephens MA (1974) EDF statistics for goodness-of-fit and some comparisons. J Am Stat Assoc 69:730–737
Stephens MA (1979) Tests of fit for the logistic distribution based on the empirical distribution function. Biometrika 66:591–595
Stute W, González-Manteiga W, Presedo-Quindimil M (1993) Bootstrap based goodness-of-fit-tests. Metrika 40:243–256
Szűcs G (2008) Parametric bootstrap tests for continuous and discrete distributions. Metrika 67:63–81
Talacko J (1956) Perks’ distributions and their role in the theory of Wiener’s stochastic variables. Trab Estad 7:159–174
Tarnopolski M (2016) Analysis of gamma-ray burst duration distribution using mixtures of skewed distributions. Mon Not R Astron Soc 458:2024–2031
Thas O (2010) Comparing distributions. Springer, New York
Vaughan DC (2002) The generalized secant hyperbolic distribution and its properties. Commun Stat Theory Methods 31:219–238
Voudouris V, Stasinopoulos D, Rigby R, Di Maio C (2011) The ACEGES laboratory for energy policy: exploring the production of crude oil. Energ Policy 39:5480–5489
Acknowledgements
I am most grateful to Dr Chris Banks and Professor Paul Garthwaite for access to the snow depth data, and to two referees and an Associate Editor for their careful reading of the original manuscript and helpful suggestions as to how it might be improved. Financial support for the research which led to the production of this paper was received from the Junta de Extremadura and the European Union in the form of grant GR15013.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pewsey, A. Parametric bootstrap edf-based goodness-of-fit testing for sinh–arcsinh distributions. TEST 27, 147–172 (2018). https://doi.org/10.1007/s11749-017-0538-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-017-0538-2
Keywords
- Anderson–Darling statistic
- Logistic distribution
- Normal distribution
- Quantile-based estimation
- Sinh–arcsinh transformation
- t-distribution