Abstract
We provide a new approach to the problem of the unique identification of distributions with a continuous density by a single regression function of order statistics or record values or, more generally, generalized order statistics. Using their Markov property we show that the characterization is unique if and only if the corresponding system of differential equations has the unique solution. This result is new even in the particular case of ordinary order statistics. This approach provides a new proof of characterization of power, exponential and Pareto distributions by linearity of corresponding regression. It also yields new examples of characterizations of distributions.
Similar content being viewed by others
References
Balakrishnan, N., Cramer, E.: The Art of Progressive Censoring. Statistics for Industry and Technology. Springer, New York (2014)
Bieniek, M.: On characterizations of distributions by regression of adjacent generalized order statistics. Metrika 66(2), 233–242 (2007)
Bieniek, M.: On characterizations of distributions by regression of generalized order statistics. Aust. N. Z. J. Stat. 51(1), 89–99 (2009)
Bieniek, M., Szynal, D.: Characterizations of distributions via linearity of regression of generalized order statistics. Metrika 58(3), 259–271 (2003)
Cramer, E., Kamps, U.: Marginal distributions of sequential and generalized order statistics. Metrika 58(3), 293–310 (2003)
Cramer, E., Tran, T.T.H.: Generalized order statistics from arbitrary distributions and the Markov chain property. J. Stat. Plan. Inference 139(12), 4064–4071 (2009)
Cramer, E., Kamps, U., Keseling, C.: Characterizations via linear regression of ordered random variables: a unifying approach. Commun. Stat. Theory Methods 33(11–12), 2885–2911 (2004)
Dembińska, A., Wesołowski, J.: Linearity of regression for non-adjacent order statistics. Metrika 48(3), 215–222 (1998)
Dembińska, A., Wesołowski, J.: Linearity of regression for non-adjacent record values. J. Stat. Plan. Inference 90(2), 195–205 (2000)
Ferguson, T.S.: On characterizing distributions by properties of order statistics. Sankhyā Ser. A 29, 265–278 (1967)
Franco, M., Ruiz, J.M.: On characterization of continuous distributions with adjacent order statistics. Statistics 26(4), 375–385 (1995)
Franco, M., Ruiz, J.M.: On characterization of continuous distributions by conditional expectation of record values. Sankhyā Ser. A 58(1), 135–141 (1996)
Franco, M., Ruiz, J.M.: Characterization based on conditional expectations of adjacent order statistics: a unified approach. Proc. Am. Math. Soc. 127(3), 861–874 (1999)
Galambos, J., Kotz, S.: Characterizations of Probability Distributions. Lecture Notes in Mathematics. Springer, New York (1978)
Kamps, U.: A concept of generalized order statistics. J. Stat. Plan. Inference 48(1), 1–23 (1995)
Keseling, C.: Conditional distributions of generalized order statistics and some characterizations. Metrika 49(1), 27–40 (1999)
Mathai, A.M.: A Handbook of Generalized Special Functions for Statistical and Physical Sciences. Oxford Science Publications, New York (1993)
Polyanin, A.D., Zaitsev, V.F.: Handbook of Exact Solutions for Ordinary Differential Equations, 2nd edn. Chapman & Hall/CRC, Boca Raton (2003)
Rao, C.R., Shanbhag, D.N.: Recent results on characterization of probability distributions: a unified approach through extensions of Deny’s theorem. Adv. Appl. Probab. 18(3), 660–678 (1986)
Acknowledgements
The authors are very grateful to anonymous referees for many valuable comments and remarks which were very helpful during the preparation of the final version of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Mariusz Bieniek was supported by Polish National Science Center under Grant No. 2015/19/B/ST1/03100.
Rights and permissions
About this article
Cite this article
Bieniek, M., Macia̧g, K. Uniqueness of characterization of absolutely continuous distributions by regressions of generalized order statistics. AStA Adv Stat Anal 102, 359–380 (2018). https://doi.org/10.1007/s10182-017-0310-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10182-017-0310-7