Circular data analysis with wrapped Poisson–Bilal distribution: Properties and applications

Abstract

This article introduces the wrapped Poisson-Bilal distribution, a novel discrete model for analyzing circular data, which is produced by wrapping the discrete Poisson-Bilal distribution on the unit circle. The probability mass function of the proposed model can be expressed as a mixture of two wrapped geometric distributions, offering a simpler and more interpretable model for circular data. We provide closed-form formulas for essential distributional properties, such as characteristic function, trigonometric moments, and circular measures of location and dispersion, allowing this model to be readily utilized in a variety of applications. The parameter of the proposed model is estimated using two methods: the maximum likelihood estimation and the moment method. The effectiveness of the model was evaluated through simulation studies in various parameter settings. Furthermore, we applied the model to three real-world datasets, demonstrating its superior performance compared to existing models based on different information criteria. These results demonstrate the capacity of the model as a powerful tool for analyzing circular data.

Keywords

• Wrapped distribution
• Poisson-Bilal distribution
• characteristic function
• trigonometric moments
• parameter estimation
• simulation study
• real data applications

References

1. [1] E. Zinhom, M. M. Nassar, S. S. Radwan and A. Elmasry, The wrapped XLindley distribution, Environ. Ecol. Stat. 30, 669–686, 2023.
2. [2] N. I. Fisher, Statistical Analysis of Circular Data, Cambridge University Press, Cambridge, 1993.
3. [3] N. I. Fisher, T. Lewis and B. J. J. Embleton, Statistical Analysis of Spherical Data, Cambridge University Press, Cambridge, 1987.
4. [4] K. V. Mardia and P. E. Jupp, Directional Statistics, Wiley, Chichester, 2000.
5. [5] C. Ley and T. Verdebout, Modern Directional Statistics, Chapman and Hall/CRC, Boca Raton, 2017.
6. [6] C. Ley and T. Verdebout, Applied Directional Statistics: Modern Methods and Case Studies, Chapman and Hall/CRC, Boca Raton, 2018.
7. [7] K. Bharath and D. Dey, Statistics on non-Euclidean spaces and manifolds [Special issue], Sankhya A 81, 2019.
8. [8] A. Pewsey and E. García-Portugués, Recent advances in directional statistics, Test 30, 1–58, 2021.

9. [9] K. V. Mardia, Comments on: Recent advances in directional statistics, Test 30, 59–63, 2021.
10. [10] M. A. S. Adnan and S. Roy,Wrapped variance gamma distribution with an application to wind direction, J. Environ. Stat. 6, 1–10, 2014.
11. [11] S. R. Jammalamadaka and A. Sengupta, Topics in Circular Statistics,World Scientific Publishing Co. Pte. Ltd., Singapore, 2001.
12. [12] S. R. Jammalamadaka and T. J. Kozubowski, New families of wrapped distributions for modeling skew-circular data, Commun. Stat. Theory Methods 33, 2059–2074, 2004.
13. [13] K. K. Jose and J. Varghese, Wrapped log Kumaraswamy distribution and its applications, Int. J. Comput. Math. 6, 1924–1930, 2018.
14. [14] S. Joshi, K. K. Jose and D. Bhati, Estimation of a change point in the hazard rate of Lindley model under right censoring, Commun. Stat. Simul. Comput. 46, 3563–3574, 2017.
15. [15] P. Lévy, Addition des variables aléatoires définies sur une circonférence, Bull. Soc. Math. France 67, 1–41, 1939.
16. [16] A. V. D. Rao, S. V. S. Girija and V. J. Devaraaj, On characteristics of wrapped gamma distribution, Eng. Sci. Technol. Int. J. 3 (2), 2013.
17. [17] A. V. D. Rao, I. R. Sarma and S. V. S. Girija, On wrapped version of some life testing models, Commun. Stat. Theory Methods 36, 2027–2035, 2007.
18. [18] G. V. L. N. Srihari, S. V.S . Girija and A. V. D. Rao, On discrete wrapped exponential distribution characteristics, Int. J. Sci. Res. Math. Stat. Sci. 5, 57–64, 2018.
19. [19] G. V. L. N. Srihari, S. V. S. Girija and A. V. D. Rao, On wrapped negative binomial model, Int. J. Math. Arch. 9, 148–153, 2018.
20. [20] T. Imoto, G. S. Shieh and K. Shimizu, Discrete circular distributions with applications to shared orthologs of paired circular genomes, Comput. Model. Eng. Sci. 123 (3), 1131–1149, 2020.
21. [21] W. Bell and S. Nadarajah, A review of wrapped distributions for circular data, Mathematics 12 (16), 2440, 2024.
22. [22] K. V. Mardia and K. Sriram, Families of discrete circular distributions with some novel applications, Sankhya A 85 (1), 1–42, 2023.
23. [23] K. Jayakumar and T. Sajayan, Wrapped generalized geometric stable distributions with an application to wind direction, Far East J. Theor. Stat. 66, 147–166, 2022.
24. [24] B. C. Arnold and A. SenGupta, Wrapping generalized Laplace and generalized discrete Laplace distributions, Axioms 14 (7), 529, 2025.
25. [25] K. Li, F. Pfaff and U. D. Hanebeck, Circular discrete reapproximation, in: Proc. 25th Int. Conf. Information Fusion (FUSION), IEEE, 1–8, 2022.
26. [26] K. Jayakumar and S. Jacob, Wrapped skew Laplace distribution on integers: A new probability model for circular data, Open J. Stat. 2, 106–114, 2012.
27. [27] S. V. S. Girija, A. V. D. Rao and G. V. L. N. Srihari, On wrapped binomial model characteristics, Math. Stat. 2, 231–234, 2014.
28. [28] E. Altun, A new one-parameter discrete distribution with associated regression and integer-valued autoregressive models, Math. Slovaca 70 (4), 979–994, 2020.
29. [29] C. Chesneau, H. S. Bakouch and T. Hussain, Modeling and applications to circular data with a wrapped Poisson–Lindley model, J. Math. Study 55 (2), 139–157, 2022.
30. [30] H. Akaike, A new look at the statistical model identification, IEEE Trans. Autom. Control 19 (6), 716–723, 2003.
31. [31] G. Schwarz, Estimating the dimension of a model, Ann. Stat. 6, 461–464, 1978.
32. [32] E. J. Hannan and B. G. Quinn, The determination of the order of an autoregression, J. R. Stat. Soc. Ser. B 41 (2), 190–195, 1979.
33. [33] H. Bozdogan, Model selection and Akaike’s information criterion (AIC): The general theory and its analytical extensions, Psychometrika 52 (3), 345–370, 1987.
34. [34] A. Thavaneswaran, S. Mandal and D. Pathmanathan, Estimation for wrapped zero inflated Poisson and wrapped Poisson distributions, Int. J. Stat. Probab. 5 (3), 2016.
35. [35] S. Jacob and K. Jayakumar, Wrapped geometric distribution: A new probability model for circular data, J. Stat. Theory Appl. 12 (4), 348–355, 2013.