Abstract
Three functional measures of the shape of univariate distributions are proposed which are consistent with respect to the convex transform order. The first two are weighted tail indices that characterize location-scale families of distributions, whilst the third is a skewness measure. Properties of the new measures are established for various classes of symmetric and asymmetric distributions, and the generalized Pareto distribution characterized in terms of them. Kernel density based estimation of the measures is also considered, and the use of the estimated functionals is illustrated in the analysis of two real data sets.
Similar content being viewed by others
References
Arnold BC, Groeneveld RA (1995) Measuring skewness with respect to the mode. Am Stat 49:34–38
Arriaza A, Di Crescenzo A, Sordo M, Suárez-Llorens A (2017) An R script to compute the weighted right/left tail index and a new skewness functional for univariate probability distributions. University of Cádiz repository, Rodin. http://hdl.handle.net/10498/19833. Accessed 16 Nov 2017
Azzalini A (1985) A class of distributions which includes the normal ones. Scand J Stat 12:171–178
Azzalini A (1986) Further results on a class of distributions which includes the normal ones. Statistica 46:199–208
Balanda KP, MacGillivray HL (1988) Kurtosis: a critical review. Am Stat 42:111–119
Barlow RE, Proschan F (1975) Statistical theory of reliability and life testing: probability models. Holt, Rinehart and Winston, New York
Belzunce F, Mulero J, Ruíz JM, Suárez-Llorens A (2015) On relative skewness for multivariate distributions. Test 24:813–834
Benjamini Y, Krieger AM (1996) Concepts and measures for skewness with data-analytic implications. Can J Stat 24:131–140
Boshnakov GN (2007) Some measures for asymmetry of distributions. Stat Probab Lett 77:1111–1116
Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press, New York
Case KE, Fair RC (1999) Principles of economics, 5th edn. Prentice-Hall, London
Charlier CVL (1905) Über das Fehlergesetz. Arkiv für Matematik, Astronomi och Fysik 8:1–9
Critchley F, Jones MC (2008) Asymmetry and gradient asymmetry functions: density-based skewness and kurtosis. Scand J Stat 35:415–437
DeCarlo LT (1997) On the meaning and use of kurtosis. Psychol Methods 2:292–307
Doksum KA (1975) Measures of location and asymmetry. Scand J Stat 2:11–22
Efron B (1979) Bootstrap methods: another look at the jackknife. Ann Stat 7:1–26
Groeneveld RA (1998) A class of quantile measures for kurtosis. Am Stat 52:994–1011
Groeneveld RA, Meeden G (1984) Measuring skewness and kurtosis. J R Stat Soc Ser D (The Statistician) 33:391–399
Hosking JRM, Wallis JR (1987) Parameter and quantile estimation for the generalized Pareto distribution. Technometrics 29:339–349
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
MacGillivray HL (1986) Skewness and asymmetry: measures and orderings. Ann Stat 14:994–1011
MacGillivray HL (1992) Shape properties of the \(g\)-and-\(h\) and Johnson families. Commun Stat Theory Methods 21:1233–1250
MacGillivray HL, Balanda KP (1988) The relationships between skewness and kurtosis. Aust J Stat 30:319–337
Oja H (1981) On location, scale, skewness and kurtosis of univariate distributions. Scand J Stat 8:154–168
Pearson K (1895) Contributions to the mathematical theory of evolution. II. Skew variation in homogeneous material. Philos Trans R Soc Lond Math Phys Eng Sci 186:343–414
Pewsey A (2015) Discussion of On families of distributions with shape parameters. Int Stat Rev 83:211–217
Pickands J (1975) Statistical inference using extreme order statistics. Ann Stat 3:119–131
Shaked M, Shanthikumar G (2007) Stochastic orders. Springer, New York
Silverman BW (1986) Density estimation for statistics and data analysis. Chapman & Hall, London
Sordo MA, Souza MC, Suárez-Llorens A (2015) A new variability order based on tail-heaviness. Statistics 49:1042–1061
Sunoj SM, Linu MN (2012) Dynamic cumulative residual Rényi’s entropy. Statistics 46:41–56
van Zwet WR (1964) Convex transformations of random variables. Mathematical centre tracts, vol 7. Mathematisch Centrum, Amsterdam
Wand MP, Jones MC (1995) Kernel smoothing. Chapman & Hall, London
Westfall PH (2014) Kurtosis as peakedness, 1905–2014. R.I.P. Am Stat 68:191–195
Yule GW (1912) An introduction to the theory of statistics, 2nd edn. Griffin, London
Zarezadeh S, Asadi M (2010) Results on residual Rényi entropy of order statistics and record values. Inf Sci 180:4195–4206
Acknowledgements
The authors would like to thank two anonymous referees and the Editor of the journal for their extensive and helpful comments which have led to significant improvements in both the content and presentation of the paper. The authors also acknowledge support received from the Ministerio de Economía y Competitividad (Spain) under Grants MTM2014-57559-P and MTM2017-89577-P, and from the group GNCS of INdAM.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all the authors, the corresponding author states that there is no conflict of interest.
Appendix A: Proofs
Appendix A: Proofs
Proof of Proposition 2.1
From (2) we observe that
where the third equivalence follows by choosing \(t=x_u\), \(s=x_{p}\) with \(u,\, p\in (0,1)\) and using the fact that \(G^{-1}F(\cdot )\) maps the quantiles of X to the corresponding quantiles of Y. \(\square \)
Proof of Corollary 2.2
Using \(F^{-1}(U)=_{st} X\), where U denotes a uniform random variable on (0, 1), we see that \((x_p-x_u)^+ f(x_p)\) and \((x_p-x_u)^- f(x_p)\) can be thought of as realizations of the random variables \((F^{-1}(U)-x_u)^+f(F^{-1}(U))\) and \((F^{-1}(U)-x_u)^-f(F^{-1}(U))\), respectively. Analogously for Y. The proof then follows directly from Corollary 2.1 and Theorem 1.A.1. in Shaked and Shanthikumar (2007). \(\square \)
Proof of Lemma 2.2
The proof follows directly from Definition 2.1 and the facts that \(-\,x_u\) is the \((1-u)\)th quantile of \(-\,X\), \(a^+=(-\,a)^-\) for all \(a\in \mathbb R\) and \(f_{-X}(-X)=f_X (X)\). \(\square \)
Proof of Lemma 2.3
From (3) we can write
The result follows on taking derivatives and using \(x'_u=1/f(x_u)\). \(\square \)
Proof of Theorem 2.1
The sufficient condition of (i) for \(R_X(u)\) follows from (3) and the fact that \((x_p-x_u)^+ f(x_p)\), with \(u,p\in (0,1)\), is not affected by translations and positive changes of scale. To prove the necessary condition, assume that \(R_X(u)=R_Y(u)\), for all \( u\in (0,1)\). Taking derivatives and using Lemma 2.3 we obtain that
which trivially implies that there exists an \(a>0\) such that
By taking derivatives again, we have \(f(x_u)=ag(y_u)\), for all \(u\in (0,1)\) or, equivalently, \(y'_u=ax'_u\), for all \(u\in (0,1)\). Therefore, there exists a \(b\in \mathbb {R}\) such that \(y_u=ax_u+b\), for all \(u\in (0,1)\), which means that \(Y=_{st}aX+b\), i.e., \(G\in [F]\). The proof for \(L_X(u)\) follows straightforwardly using Lemma 2.2.
Part (ii) holds because of the constructions of \(R_X(u)\) and \(L_X(u)\).
In order to prove (iii), assume that X and Y are symmetric and \(X\le _{k}Y\), which means that \(G^{-1}F(x)\) is convex for all \(x>m_X\), where \(m_X\) denotes the median of X. This implies, by restricting Proposition 2.1 to \(D=(m_X,+\,\infty )\), that \(R_X(u) \ge R_Y(u)\), for all \(u \in (1/2,1)\). Now, from Lemma 2.2, the location invariance property and the symmetry of X we have
and it follows that \(L_X(u) \ge L_Y(u)\), for all \(u\in (0,1/2)\). \(\square \)
Proof of Theorem 2.2
From the assumptions, and using L’Hôpital’s rule and Lemma 2.3, we obtain
where the second equality follows by changing the variables in the integrals. It follows that there exists an \(u_0 \in (0,1)\) such that
for all \(u\ge u_0\). \(\square \)
Proof of Proposition 3.1
The necessary condition follows from Theorem 2.1(i). In order to prove the sufficient condition, let X be a GP(c, k) random variable with \(c < 2\) and \(k > 0\). Observe that, if \(c=0,\), X follows an exponential distribution and the result follows from Table 1. Suppose, therefore, that \(c\ne 0\). From (5) the density and quantile functions of X are given by
and therefore
\(\square \)
Proof of Proposition 3.2
Let f denote the density of X. From Lemma 2.3 we have
Then, there exists a \(c>0\) such that
Using \(f(x_u)=1/x_u^{\prime }\), we have
and by integration we obtain
which concludes the proof. \(\square \)
Proof of Proposition 4.1
Part (i) follows from Theorem 2.1(i). Similarly, (iii) and (iv) follow straightforwardly from Lemma 2.2 and Theorem 2.1(ii). To prove (ii), first suppose that X is symmetric about the median \(m_X\), which implies that \(X-m_X=_{st} -X+m_X\). Hence, using Lemma 2.2 and Theorem 2.1(i), we have
which implies that \(\textit{SK}_X(u) = 0\), for all \(u\in (0,1)\).
Conversely, assume \(\textit{SK}_X(u) = 0\), for all \(u\in (0,1)\). From the definition of \(\textit{SK}_X(u)\), this means that \(L_{X}(u)=R_{X}(1-u)\) holds for \(u\in (0,1)\). It follows from Lemma 2.2 that \(R_{-X}(1-u)=R_{X}(1-u)\) for \(u\in (0,1)\). Applying Theorem 2.1(i), there exists an \(a>0\) and a \(b\in \mathbb {R}\) such that \(-\,X=_{st}aX+b\). By taking variances and expectations it follows that \(a=1\) and \(b=-2 \mu _X\), where \(\mu _X\) represents the expectation of X. Therefore \(-\,X+\mu _X=_{st} X -\mu _X\) which implies that X is symmetrically distributed. \(\square \)
Proof of Proposition 4.2
Suppose that \(-\,X\le _c X\). The proof for when \(X\le _c -X\) is analogous. It follows from Proposition 4.1(iv) that \(\textit{SK}_{-X}(u)\le \textit{SK}_X(u)\), for all \(u\in (0,1)\), which implies, by Proposition 4.1(iii), that \(-\,\textit{SK}_{X}(u)\le \textit{SK}_X(u)\) holds for all \(u\in (0,1)\). This concludes the proof. \(\square \)
Proof of Corollary 4.1
Assume that f is increasing on \(D_X\). The proof for when f is decreasing is similar. From Proposition 4.2 it is sufficient to verify that \(X \le _{c} -X\). Denoting the distribution function of \(-\,X\) by \(G(x)=1-F(-x)\), we have
where the second equality follows by taking derivatives and the third one by making a change of variable. Since f is increasing, we see that \(f(x_u)\) and \(f(x_{1-u})\) are increasing and decreasing functions of u, respectively, in the interval (0, 1). This concludes the proof. \(\square \)
Rights and permissions
About this article
Cite this article
Arriaza, A., Di Crescenzo, A., Sordo, M.A. et al. Shape measures based on the convex transform order. Metrika 82, 99–124 (2019). https://doi.org/10.1007/s00184-018-0667-y
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00184-018-0667-y