×

Kurtosis test of modality for rotationally symmetric distributions on hyperspheres. (English) Zbl 1440.62202

Summary: A test of modality of rotationally symmetric distributions on hyperspheres is proposed. The test is based on a modified multivariate kurtosis defined for directional data on \(\mathbb{S}^d\). We first reveal a relationship between the multivariate kurtosis and the types of modality for Euclidean data. In particular, the kurtosis of a rotationally symmetric distribution with decreasing sectional density is greater than the kurtosis of the uniform distribution, while the kurtosis of that with increasing sectional density is less. For directional data, we show an asymptotic normality of the modified spherical kurtosis, based on which a large-sample test is proposed. The proposed test of modality is applied to the problem of preventing overfitting in non-geodesic dimension reduction of directional data. The proposed test is superior than existing options in terms of computation times, accuracy and preventing overfitting. This is highlighted by a simulation study and two real data examples.

MSC:

62H11 Directional data; spatial statistics
62H25 Factor analysis and principal components; correspondence analysis
62E10 Characterization and structure theory of statistical distributions
62R30 Statistics on manifolds

Software:

rotasym
Full Text: DOI

References:

[1] Abbasi, R.; Abdou, Y.; Abu-Zayyad, T.; Adams, J.; Aguilar, J. A.; Ahlers, M.; Altmann, D.; Andeen, K.; Auffenberg, J.; Bai, X., Search for a diffuse flux of astrophysical muon neutrinos with the IceCube 40-string detector, Phys. Rev. D, 84, 8, Article 082001 pp. (2011)
[2] Alsmeyer, G., Nonnegativity of odd functional moments of positive random variables with decreasing density, Statist. Probab. Lett., 26, 1, 75-82 (1996) · Zbl 0848.60022
[3] Bhattacharya, R.; Lin, L., Omnibus CLTs for Fréchet means and nonparametric inference on non-Euclidean spaces, Proc. Amer. Math. Soc., 145, 1, 413-428 (2017) · Zbl 1353.60019
[4] Bingham, C.; Mardia, K. V., A small circle distribution on the sphere, Biometrika, 65, 2, 379-389 (1978) · Zbl 0388.62020
[5] Damon, J.; Marron, J., Backwards principal component analysis and principal nested relations, J. Math. Imaging Vision, 50, 1-2, 107-114 (2014) · Zbl 1308.62124
[6] Dryden, I. L.; Kim, K.-R.; Laughton, C. A.; Le, H., Principal nested shape space analysis of molecular dynamics data, Ann. Appl. Stat., 13, 4, 2213-2234 (2019) · Zbl 1435.62216
[7] Dryden, I. L.; Mardia, K. V., Statistical Shape Analysis (2016), Wiley: Wiley Chichester · Zbl 0901.62072
[8] Eltzner, B.; Huckemann, S.; Mardia, K. V., Torus principal component analysis with applications to RNA structure, Ann. Appl. Stat., 12, 2, 1332-1359 (2018) · Zbl 1405.62173
[9] Eltzner, B.; Jung, S.; Huckemann, S., Dimension reduction on polyspheres with application to skeletal representations, (International Conference on Geometric Science of Information (2015), Springer), 22-29 · Zbl 1376.94012
[10] Franceschini, C.; Loperfido, N., Some inequalities between measures of multivariate kurtosis, with application to financial returns, (Mathematical and Statistical Methods for Actuarial Sciences and Finance (2012)), 211-218 · Zbl 1238.91149
[11] García-Portugués, E.; Paindaveine, D.; Verdebout, T., On optimal tests for rotational symmetry against new classes of hyperspherical distributions, J. Amer. Stat. Assoc. (2020), in press · Zbl 1452.62362
[12] Hildebrand, D. K., Kurtosis measures bimodality?, Amer. Statist., 25, 1, 42-43 (1971)
[13] Hong, J.; Vicory, J.; Schulz, J.; Styner, M.; Marron, J. S.; Pizer, S. M., Non-Euclidean classification of medically imaged objects via s-reps, Med. Image Anal., 18, 31, 37-45 (2016)
[14] Huckemann, S. F.; Eltzner, B., Backward nested descriptors asymptotics with inference on stem cell differentiation, Ann. Statist., 46, 5, 1994-2019 (2018) · Zbl 1405.62070
[15] Jung, S.; Dryden, I. L.; Marron, J. S., Analysis of principal nested spheres, Biometrika, 99, 3, 551-568 (2012) · Zbl 1437.62507
[16] Jung, S.; Foskey, M.; Marron, J. S., Principal arc analysis on direct product manifolds, Ann. Appl. Stat., 5, 1, 578-603 (2011) · Zbl 1220.62077
[17] Kim, B.; Huckemann, S.; Schulz, J.; Jung, S., Small-sphere distributions for directional data with application to medical imaging, Scand. J. Stat., 46, 1047-1071 (2019) · Zbl 1444.62067
[18] Kollo, T., Multivariate skewness and kurtosis measures with an application in ICA, J. Multivariate Anal., 99, 10, 2328-2338 (2008) · Zbl 1294.62021
[19] Kollo, T.; Roos, A., On kotz-type elliptical distributions, (Contemporary Multivariate Analysis and Design of Experiments (2005), World Scientific), 159-170 · Zbl 1266.62034
[20] Koziol, J. A., An alternative formulation of Neyman’s smooth goodness of fit tests under composite alternatives, Metrika, 34, 17-24 (1987) · Zbl 0613.62054
[21] Koziol, J. A., A note on measures of multivariate kurtosis, Biom. J., 31, 619-624 (1989)
[22] Ley, C.; Swan, Y.; Thiam, B.; Verdebout, T., Optimal R-estimation of a spherical location, Statist. Sinica, 23, 1, 305-332 (2013) · Zbl 1259.62044
[23] Loperfido, N., A new kurtosis matrix, with statistical applications, Linear Algebra Appl., 512, 1-17 (2017) · Zbl 1403.62129
[24] Loperfido, N., Some remarks on Koziol’s kurtosis, J. Multivariate Anal., 175, 104565 (2020) · Zbl 1435.62218
[25] Mardia, K. V., Measures of multivariate skewness and kurtosis with applications, Biometrika, 57, 3, 519-530 (1970) · Zbl 0214.46302
[26] Mardia, K. V.; Jupp, P. E., Directional Statistics, 429 (2000), Wiley: Wiley New York · Zbl 0935.62065
[27] McClure, R. K.; Styner, M.; Maltbie, E.; Liebermann, J. A.; Gouttard, S.; Gerig, G.; Shi, X.; Zhu, H., Localized differences in caudate and hippocampal shape are associated with schizophrenia but not antipsychotic type, Psychiatry Research: Neuroimaging, 211, 1, 1-10 (2013)
[28] Mori, T. F.; Rohatgi, V. K.; Szekely, G. J., On multivariate skewness and kurtosis, Theory Probab. Appl., 38, 547-551 (1993) · Zbl 0807.60020
[29] Paindaveine, D.; Verdebout, T., Optimal rank-based tests for the location parameter of a rotationally symmetric distribution on the hypersphere, (Mathematical Statistics and Limit Theorems (2015), Springer), 249-269 · Zbl 1320.62131
[30] Paindaveine, D.; Verdebout, T., Inference on the mode of weak directional signals: a Le Cam perspective on hypothesis testing near singularities, Ann. Statist., 45, 2, 800-832 (2017) · Zbl 1371.62043
[31] Pinkus, A., Totally Positive Matrices, Cambridge Tracts in Mathematics (2009), Cambridge University Press: Cambridge University Press Cambridge · Zbl 0553.15008
[32] Pizer, S. M.; Jung, S.; Goswami, D.; Zhao, X.; Chaudhuri, R.; Damon, J. N.; Huckemann, S.; Marron, J. S., Nested sphere statistics of skeletal models, (Breus, M.; Bruckstein, A.; Maragos, P., Proc. Dagstuhl Workshop on Innovations for Shape Analysis: Models and Algorithms. Proc. Dagstuhl Workshop on Innovations for Shape Analysis: Models and Algorithms, Innovations for Shape Analysis: Models and Algorithms (2013), Springer: Springer New York), 93-115 · Zbl 1314.68357
[33] Rösler, U., Distributions slanted to the right, Stat. Neerl., 49, 1, 83-93 (1995) · Zbl 0835.62015
[34] Schulz, J.; Pizer, S. M.; Marron, J. S.; Godtliebsen, F., Non-linear hypothesis testing of geometric object properties of shapes applied to hippocampi, J. Math. Imaging Vision, 54, 15-34 (2016) · Zbl 1338.92063
[35] Shi, X.; Ibrahim, J. G.; Lieberman, J.; Styner, M.; Li, Y.; Zhu, H., Two-stage empirical likelihood for longitudinal neuroimaging data, Ann. Appl. Stat., 5, 2B, 1132-1158 (2011) · Zbl 1223.62024
[36] Shohat, J. A.; Tamarkin, J. D., The Problem of Moments (1943), American Mathematical Scociety: American Mathematical Scociety Providence · Zbl 0063.06973
[37] Westfall, P. H., Kurtosis as peakedness, 1905-2014. R.I. P., Amer. Statist., 68, 3, 191-195 (2014) · Zbl 07653656
[38] Yu, Q.; Lu, X.; Marron, J., Principal nested spheres for time-warped functional data analysis, J. Comput. Graph. Statist., 26, 1, 144-151 (2017)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.