Some characteristics of tensor-variate skew-normal distribution and its application in image analysis

Document Type : Research Paper

Authors

1 Department of Statistics, University of Birjand, Birjand, Iran

2 Department of Estadistica, Pontificia Universidad Catolica de Chile, Santiago, Chile

Abstract

In recent years, due to the increasing growth of technology and new technologies, data is obtained in more complex structures as the main component in analysis. One of these complex structures is tensors. Therefore, in order to answer this need (analysis of data with tensor structure), it is necessary to expand statistical concepts and methods in the field of data with tensor structure. On the other hand, in reality, we may also encounter skew data. Therefore, in this article, we have introduced the skew normal tensor distribution and obtained some of its important statistical properties. Subsequently, we employed the EM algorithm to obtain maximum likelihood estimates of the parameters and assessed their accuracy through simulation studies. Finally, we have shown the effectiveness of the obtained estimators with real data.

Keywords

Main Subjects

References

[1] Arashi, M. (2021). Some theoretical results on tensor elliptical distribution. Computational Statistics and Modeling, 1(2), 29{41. https://doi.org/10.48550/arXiv.1709.00801
[2] Anderlucci, L. and Viroli, C. (2015). Covariance pattern mixture models for the analysis of multivariate heterogeneous longitudinal data. The Annals of Applied Statistics, 9(2), 777{800. https://doi.org/10.1214/15-AOAS816
[3] Azzalini, A. (2013). The Skew-Normal and Related Families. Cambridge, Cambridge University Press.
[4] Basser, P.J. and Pajevic, S. (2003). A normal distribution for tensor-valued random variables: applications to di usion tensor MRI. IEEE Transactions on Medical Imaging, 22(7), 785{794. https://doi.org/10.1109/TMI.2003.815059
[5] Bi, X., Tang, X., Yuan, Y., Zhang, Y. and Qu, Annie. (2021). Tensors in Statistics. Annual Review of Statistics and Its Application, 8(1) , 345{368. http://dx.doi.org/10.1146/annurev-statistics-042720-020816
[6] Drygas, H. (1984). Linear suciency and some applications in multilinear estimation. Journal of Multivariate Analysis, 16(1), 71{84. https://doi.org/10.1016/0047-259X(85)90052-1
[7] Gallaugher, Michael P. B., Tait, Peter A. and McNicholas Paul D. (2021). Four skewed tensor distributions. Available from:arXiv:2106.08984. https://doi.org/10.48550/arXiv.2106.08984
[8] Ghannam, M. and Nkurunziza, S. (2022). The risk of tensor Stein-rules in elliptically contoured distributions. Statistics, 56(2), 421{454. https://doi.org/10.1080/02331888.2022.2051508
[9] Kolda, T.G. and Bader, B.W. (2009). Tensor decompositions and applications. SIAM Review, 51(3), 455{500. https://doi.org/10.1137/07070111X
[10] Krizhevsky, A. and Hinton, Geo rey. (2009). Learning Multiple Layers of Features from Tiny Images. Toronto, ON, Canada. https://www.cs.utoronto.ca/ kriz/learning-features-2009-TR.pdf
[11] Lee, I., Sinha, D., Mai, Q., Zhang, X. and Bandyopadhyay, D. (2023). Bayesian regression analysis of skewed tensor responses. Biometrics, 79 , 1814{1825. https://doi.org/10.1111 /biom.13743
[12] Llosa-Vite, C. and Maitra, R. (2022). Elliptically-Contoured Tensor-variate Distributions with Application to Improved Image Learning. Available from:arXiv:2211.06940. https://doi.org/10.48550/arXiv.2211.06940
[13] Northcott, D.G. (1984). Multilinear Algebra. Cambridge, Cambridge University Press. https://doi.org/10.1017/cbo9780511565939
[14] Ohlson, M., Rauf Ahmad, M. and Von Rosen, D. (2013). The multilinear normal distribution: Introduction and some basic properties. Journal of Multivariate Analysis, 113, 37{47. https://doi.org/10.1016/j.jmva.2011.05.015
[15] Srivastava, M., von Rosen, T. and von Rosen, D. (2008). Models with a Kronecker product covariance structure: Estimation and testing. Mathematical Methods of Statistics, 17, 357{370. https://doi.org/10.3103/S1066530708040066
[16] Tait, Peter A., McNicholas, Paul D. and Obeid, J. (2020). Clustering Higher Order Data: An Application to Pediatric Multi-variable Longitudinal Data. Available from:arXiv:1907.08566. https://doi.org/10.48550/arXiv.1907.08566 

Files

History

  • Receive Date: 11 February 2025
  • Revise Date: 14 May 2025
  • Accept Date: 08 July 2025

Share

How to cite

Statistics