Hostname: page-component-cd9895bd7-lnqnp Total loading time: 0 Render date: 2024-12-28T13:45:28.897Z Has data issue: false hasContentIssue false
  • English
  • Français

Theory for high-order bounds in functional principal components analysis

Published online by Cambridge University Press:  01 January 2009

PETER HALL  and
MOHAMMAD HOSSEINI–NASAB
PETER HALL
Affiliation:
Department of Mathematics and Statistics, The University of Melbourne, VIC 3010, Australia e-mail: [email protected]
MOHAMMAD HOSSEINI–NASAB
Affiliation:
Department of Statistics, Faculty of Basic Science, Tarbiatmodares University, Tehran, Iran1411713116 e-mail: [email protected]
Get access Rights & Permissions [Opens in a new window]

Abstract

Functional data analysis, or FDA, is a relatively new and rapidly growing area of statistics. A substantial part of the interest in the field derives from new types of data that are generated through the application of new technologies. Statistical methodologies, such as linear regression, which are effectively finite-dimensional in conventional statistical settings, become infinite-dimensional in the context of functional data. As a result, the convergence rates of estimators based on functional data can be relatively slow, and so there is substantial interest in methods for dimension reduction, such as principal components analysis (PCA). However, although the statistical development of PCA for FDA has been underway for approximately two decades, relatively high-order theoretical arguments have been largely absent. This makes it difficult to assess the impact that, for example, eigenvalue spacings have on properties of eigenvalue estimators, or to develop concise first-order limit theory for linear functional regression. This paper shows how to overcome these hurdles. It develops rigorous arguments that underpin stochastic expansions of estimators of eigenvalues and eigenfunctions, and shows how to use them to answer statistical questions. The theory is based on arguments from operator theory, made more challenging by the requirement of statisticians that closeness of functions be measured in the L∞, rather than L2, metric. The statistical implications of the properties we develop have been discussed elsewhere, but the theoretical arguments that lie behind them have not been presented before.

Type
Research Article
Information
Mathematical Proceedings of the Cambridge Philosophical Society , Volume 146 , Issue 1 , January 2009 , pp. 225 - 256
Copyright
Copyright © Cambridge Philosophical Society 2008

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

REFERENCES

Bhatia, R., Davis, C. and McIntosh, A.Perturbation of spectral subspaces and solution of linear operator equations. Linear Algebra Appl. 52/53 (1983), 45–67.Google Scholar
Bosq, D.Propriétés des opérateurs de covariance empiriques d'un processus stationnaire hilbertien. C. R. Acad. Sci. Paris: Sér. I 309 (1989), 873–875.Google Scholar
Bosq, D.Modelization, nonparametric estimation and prediction for continuous time processes. In Nonparametric Functional Estimation and Related Topics, (Spetses, 1990). NATO Adv. Sci. Inst. Ser. C Math. Phys. Sci. 335, 509–529 (Kluwer, 1991).Google Scholar
Bosq, D. Linear Processes in Function Spaces. Theory and Applications. Springer Lecture Notes in Statistics 149 (Springer, 2000).Google Scholar
Bosq, D. and Cheze, N.Erreur quadratique asymptotique optimale de l'estimateur non paramétrique de la régression pour des observations discrétisées d'un processus stationnaire à temps continu. C. R. Acad. Sci. Paris Sér. I 317 (1993), 891–894.Google Scholar
Bosq, D. and Cheze–Payaud, N.Optimal asymptotic quadratic error of nonparametric regression function estimates for a continuous-time process from sampled-data. Statistics 32 (1999), 229–247.Google Scholar
Cai, T. and Hall, P.Prediction in functional linear regression. Ann. Statist. 34 (2007), 2159–2179.Google Scholar
Cardot, H., Mas, A. and Sarda, P.CLT in functional linear regression models. Probab. Theory Related Fields 138 (2007), 325–361.CrossRefGoogle Scholar
Dauxois, J., Pousse, A. and Romain, Y.Asymptotic theory for the principal component analysis of a vector random function: some applications to statistical inference. J. Multivariate Anal. 12 (1982), 136–154.CrossRefGoogle Scholar
Hall, P. and Hosseini–Nasab, M.On properties of functional principal components analysis. J. R. Stat. Soc. Ser. B 68 (2006), 109–126.Google Scholar
Mas, A. and Menneteau, L. Perturbation approach applied to the asymptotic study of random operators. In High Dimensional Probability, III, Eds. Hoffmann-Joergensen, J., Marcus, M. B. and Wellner, J. A., (Birkhäuser, 2003), 127–134.Google Scholar
Ramsay, J. O. and Silverman, B. W.Applied Functional Data Analysis: Methods and Case Studies (Springer, 2002).Google Scholar
Ramsay, J. O. and Silverman, B. W. Functional Data Analysis, 2nd edn. (Springer, 2005).Google Scholar
Rice, J. A. and Silverman, B. W.Estimating the mean and covariance structure nonparametrically when the data are curves. J. Roy. Statist. Soc. Ser. B 53 (1991), 233–243.Google Scholar
Silverman, B. W.Incorporating parametric effects into functional principal components analysis. J. Roy. Statist. Soc. Ser. B 57 (1995), 673–689.Google Scholar
Silverman, B. W.Smoothed functional principal components analysis by choice of norm. J. Ann. Statist. 24 (1996), 1–24.Google Scholar