Abstract
We propose a methodology for testing two-sample means in high-dimensional functional data that requires no decaying pattern on eigenvalues of the functional data. To the best of our knowledge, we are the first to consider such a problem and address it. To be specific, we devise a confidence region for the mean curve difference between two samples, which directly establishes a rigorous inferential procedure based on the multiplier bootstrap. In addition, the proposed test permits the functional observations in each sample to have mutually different distributions and arbitrary correlation structures, which is regarded as the desired property of distribution/correlation-free, leading to a more challenging scenario for theoretical development. Other desired properties include the allowance for highly unequal sample sizes, exponentially growing data dimension in sample sizes and consistent power behavior under fairly general alternatives. The proposed test is shown uniformly convergent to the prescribed significance, and its finite sample performance is evaluated via the simulation study and an implementation to electroencephalography data.
References
Bai Z D, Saranadasa H. Effect of high dimension: By an example of a two sample problem. Statist Sinica, 1996, 6: 311–329
Cai T T, Hall P. Prediction in functional linear regression. Ann Statist, 2006, 34: 2159–2179
Cai T T, Liu W D, Xia Y. Two-sample test of high dimensional means under dependence. J R Stat Soc Ser B Stat Methodol, 2014, 76: 349–372
Chang J Y, Zheng C, Zhou W-X, et al. Simulation-based hypothesis testing of high dimensional means under covariance heterogeneity. Biometrics, 2017, 73: 1300–1310
Chen S X, Qin Y-L. A two-sample test for high-dimensional data with applications to gene-set testing. Ann Statist, 2010, 38: 808–835
Fan Y Y, James G M, Radchenko P. Functional additive regression. Ann Statist, 2015, 43: 2296–2325
Gregory K B, Carroll R J, Baladandayuthapani V, et al. A two-sample test for equality of means in high dimension. J Amer Statist Assoc, 2015, 110: 837–849
Hall P, Horowitz J L. Methodology and convergence rates for functional linear regression. Ann Statist, 2007, 35: 70–91
Hall P, Hosseini-Nasab M. On properties of functional principal components analysis. J R Stat Soc Ser B Stat Methodol, 2006, 68: 109–126
Hall P, van Keilegom I. Two-sample tests in functional data analysis starting from discrete data. Statist Sinica, 2007, 17: 1511–1531
Hussain L, Aziz W, Nadeem S A, et al. Electroencephalography (EEG) analysis of alcoholic and control subjects using multiscale permutation entropy. J Multidiscip Engrg Sci Technol, 2014, 1: 380–387
Kong D H, Xue K J, Yao F, et al. Partially functional linear regression in high dimensions. Biometrika, 2016, 103: 147–159
Krzyśko M, Smaga L. Two-sample tests for functional data using characteristic functions. Austrian J Statist, 2021, 50: 53–64
Lee J S, Cox D D, Follen M. A two sample test for functional data. Comm Statist Appl Methods, 2015, 22: 121–135
Lin Z H, Lopes M E, Müller H-G. High-dimensional MANOVA via bootstrapping and its application to functional and sparse count data. J Amer Statist Assoc, 2023, in press
Pomann G-M, Staicu A-M, Ghosh S. A two-sample distribution-free test for functional data with application to a diffusion tensor imaging study of multiple sclerosis. J R Stat Soc Ser C Appl Stat, 2016, 65: 395–414
Reiss P T, Ogden R T. Functional principal component regression and functional partial least squares. J Amer Statist Assoc, 2007, 102: 984–996
Rice J A, Silverman B W. Estimating the mean and covariance structure nonparametrically when the data are curves. J R Stat Soc Ser B Stat Methodol, 1991, 53: 233–243
Srivastava M S, Kubokawa T. Tests for multivariate analysis of variance in high dimension under non-normality. J Multivariate Anal, 2013, 115: 204–216
Wang Q Y. Two-sample inference for sparse functional data. Electron J Stat, 2021, 15: 1395–1423
Xue K J, Yao F. Distribution and correlation-free two-sample test of high-dimensional means. Ann Statist, 2020, 48: 1304–1328
Xue K J, Yao F. Hypothesis testing in large-scale functional linear regression. Statistica Sinica, 2021, 31: 1101–1123
Yao F, Müller H-G, Wang J-L. Functional linear regression analysis for longitudinal data. Ann Statist, 2005, 33: 2873–2903
Zhang C Q, Peng H, Zhang J-T. Two samples tests for functional data. Comm Statist Theory Methods, 2010, 39: 559–578
Acknowledgements
This work was supported by National Natural Science Foundation of China (Grant No. 11901313), Fundamental Research Funds for the Central Universities, Key Laboratory for Medical Data Analysis and Statistical Research of Tianjin, and Key Laboratory of Pure Mathematics and Combinatorics, Ministry of Education. The author thanks the referees for their insightful comments.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Xue, K. Distribution/correlation-free test for two-sample means in high-dimensional functional data with eigenvalue decay relaxed. Sci. China Math. 66, 2337–2346 (2023). https://doi.org/10.1007/s11425-022-2042-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-022-2042-6
Keywords
- high dimension
- functional data
- eigenvalue decay relaxed
- multiplier bootstrap
- distribution/correlation-free