ABSTRACT

The unified weighing scheme for the local-linear smoother in analysing functional data can deal with data that are dense, sparse or of neither type. In this paper, we focus on the convergence rate of functional principal component analysis using this method. Almost sure asymptotic consistency and rates of convergence for the estimators of eigenvalues and eigenfunctions have been established. We also provide the convergence rate of the variance estimation of the measurement error. Based on the results, the number of observations within each curve can be of any rate relative to the sample size, which is consistent with the earlier conclusions about the asymptotic properties of the mean and covariance estimators.

References

1. Fan, J. Q., & Gijbels, I. (1996). Local polynomial modelling and its applications. London: Chapman and Hall. [Google Scholar]
2. Fan, J. Q., & Zhang, W. Y. (2000). Simultaneous confidence bands and hypothesis testing in varying-coefficient models. Scandinavian Journal of Statistics, 27, 715–731. doi: 10.1111/1467-9469.00218 [Crossref], [Web of Science ®], [Google Scholar]
3. Ferraty, F., & Vieu, P. (2006). Nonparametric functional data analysis: Theory and practice. Berlin: Springer. [Google Scholar]
4. Greven, S., Crainiceanu, C. M., Caffo, B. S., & Reich, D. (2010). Longitudinal functional principal component analysis. Electronic Journal of Statistics, 4, 1022–1054. doi: 10.1214/10-EJS575 [Crossref], [Web of Science ®], [Google Scholar]
5. Hall, P., & Hosseini-Nasab, M. (2006). On properties of functional principal components analysis. Journal of the Royal Statistical Society, Series B, 68, 109–126. doi: 10.1111/j.1467-9868.2005.00535.x [Crossref], [Google Scholar]
6. Hall, P., Müller, H.-G., & Wang, J.-L. (2006). Properties of principal component methods for functional and longitudinal data analysis. Annals of Statistics, 34, 1493–1517. doi: 10.1214/009053606000000272 [Crossref], [Web of Science ®], [Google Scholar]
7. James, G. M., Hastie, T. J., & Sugar, C. A. (2000). Principal component models for sparse functional data. Biometrika, 87, 587–602. doi: 10.1093/biomet/87.3.587 [Crossref], [Web of Science ®], [Google Scholar]
8. Jiang, C. R., & Wang, J. -L. (2010). Covariate adjusted functional principal components analysis for longitudinal data. Annals of Statistics, 38, 1194–1226. doi: 10.1214/09-AOS742 [Crossref], [Web of Science ®], [Google Scholar]
9. Li, Y. H., & Hsing, T. (2010). Uniform convergence rates for nonparametric regression and principal component analysis in functional/longitudinal data. Annals of Statistics, 38, 3321–3351. doi: 10.1214/10-AOS813 [Crossref], [Web of Science ®], [Google Scholar]
10. Li, Y., Wang, N., & Carroll, R. J. (2013). Selecting the number of principal components in functional data. Journal of the American Statistical Association, 108, 1284–1294. doi: 10.1080/01621459.2013.788980 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
11. Ramsay, J. O., & Silverman, B. W. (2005). Functional data analysis. New York: Springer. [Crossref], [Google Scholar]
12. Rice, J. A., & Silverman, B. W. (1991). Estimating the mean and covariance structure nonparametrically when the data are curves. Journal of the Royal Statistical Society, Series B, 53, 233–243. [Google Scholar]
13. Staniswalis, J. G., & Lee, J. J. (1998). Nonparametric regression analysis of longitudinal data. Journal of the American Statistical Association, 93, 1403–1418. doi: 10.1080/01621459.1998.10473801 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
14. Yao, F., & Lee, T. C. M. (2006). Penalized spline models for functional principal component analysis. Journal of the Royal Statistical Society, Series B, 68, 3–25. doi: 10.1111/j.1467-9868.2005.00530.x [Crossref], [Google Scholar]
15. Yao, F., Müller, H.-G., & Wang, J.-L. (2005). Functional data analysis for sparse longitudinal data. Journal of the American Statistical Association, 100, 577–590. doi: 10.1198/016214504000001745 [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
16. Zhang, X. K., & Wang, J. -L. (2016). From sparse to dense functional data and beyond. Annals of Statistics, 44, 2281–2321. doi: 10.1214/16-AOS1446 [Crossref], [Web of Science ®], [Google Scholar]
17. Zhou, L., Lin, H. Z., & Liang, H (2017). Efficient estimation of the nonparametric mean and covariance functions for longitudinal and sparse functional data. Journal of the American Statistical Association. doi: 10.1080/01621459.2017.1356317. [Taylor & Francis Online], [Google Scholar]
18. Zhu, H. T., Li, R. Z., & Kong, L. L. (2012). Multivariate varying coefficient model for functional responses. Annals of Statistics, 40, 2634–2666. doi: 10.1214/12-AOS1045 [Crossref], [Web of Science ®], [Google Scholar]

Articles from other publishers

Xingyu Yan, Yingchun Zhou, Xiaolong Pu, Peng Zhao. (2021) Functional Multiple-Outcome Model in Application to Multivariate Growth Curves of Infant Data. Journal of Systems Science and Complexity 24. 
Crossref

Ziqi Chen, Jianhua Hu, Hongtu Zhu. (2020) Surface functional models. Journal of Multivariate Analysis 180, pages 104664. 
Crossref