Abstract
Symmetric positive definite (SPD) matrices arise in a wide range of applications including diffusion tensor imaging (DTI), cosmic background radiation, and as covariance matrices. A complication when working with such data is that the space of SPD matrices is a manifold, so traditional statistical methods may not be directly applied. However, there are nonparametric procedures based on resampling for statistical inference for such data, but these can be slow and computationally tedious. Schwartzman (Int Stat Rev 84(3):456–486, 2016). introduced a lognormal distribution on the space of SPD matrices, providing a convenient framework for parametric inference on this space. Our goal is to check how robust confidence regions based on this distributional assumption are to a lack of lognormality. The methods are illustrated in a simulation study by examining the coverage probability of various mixtures of distributions.
Similar content being viewed by others
References
Bhattacharya, R.N., Patrangenaru, V.: Large sample theory of intrinsic and extrinsic sample means on manifolds-Part I. Ann. Stat. 31(1), 1–29 (2003)
Bhattacharya, R.N., Patrangenaru, V.: Large sample theory of intrinsic and extrinsic sample means on manifolds-Part II. Ann. Stat. 33(3), 1225–1259 (2005)
Bhattacharya, R.N., Ellingson, L., Liu, X., Patrangenaru, V., Crane, M.: Extrinsic analysis on manifolds is computationally faster than intrinsic analysis with application to quality control by machine vision. Appl. Stoch. Models Bus. Ind. 28, 222–235 (2012)
Dryden, I.L., Koloydenko, A., Zhou, D.: Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imaging. Ann. Appl. Stat. 3(3), 1102–1123 (2017)
Ellingson, L., Groisser, D., Osborne, D.E., Patrangenaru, V., Schwartzman, A.: Nonparametric bootstrap of sample means of positive definite matrices with an application to diffusion-tensor-imaging data. Commun. Stat. Simul. Comput. 46(6), 4851–4879 (2016)
Fisher, R. A.: Dispersion on a sphere. Proc. R. Soc. Lond. A 217295–217305 (1953). https://doi.org/10.1098/rspa.1953.0064
Hendriks, H., Landsman, Z.: Mean location and sample mean location on manifolds: asymptotics, tests, confidence regions. J. Multivar. Anal. 67(2), 227–243 (1998)
Osborne, D.E., Patrangenaru, V., Ellingson, L., Groisser, D., Schwartzman, A.: Nonparametric two-sample tests on homogeneous riemannian manifolds, cholesky decompositions and diffusion tensor image analysis. J. Multivar. Anal. 119, 163–175 (2013)
Schwartzman, A.: Lognormal distributions and geometric averages of positive definite matrices. Int. Stat. Rev. 84(3), 456–486 (2016). https://doi.org/10.1111/insr.12113
Watson, G.S., Williams, E.J.: On the Construction of significance tests on the circle and the sphere. Biometrika 43(3–4), 344–352 (1956). https://doi.org/10.1093/biomet/43.3-4.344
Acknowledgements
The authors are grateful to the anonymous referees and associate editors for their many valuable comments and suggestions that improved the manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ahanda, B., Osborne, D.E. & Ellingson, L. Robustness of lognormal confidence regions for means of symmetric positive definite matrices when applied to mixtures of lognormal distributions. METRON 80, 281–303 (2022). https://doi.org/10.1007/s40300-022-00234-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40300-022-00234-z