×

Robust functional principal components for irregularly spaced longitudinal data. (English) Zbl 1478.62152

Summary: Consider longitudinal data \(x_{ij}\), with \(i=1,\dots,n\) and \(j=1, \dots, p\), where \(x_{ij}\) is the observation of the smooth random function \(X_i(\cdot)\) at time \(t_j\). The goal of this paper is to develop a parsimonious representation of the data by a linear combination of a set of \(q<p\) smooth functions \(H_k(\cdot)\) (\(k=1, \dots, q\)) in the sense that \(x_{ij}\approx \mu_j+\sum_{k=1}^q\beta_{ki}H_k\left( t_j\right)\). This representation should be resistant to atypical \(X_i\)’s (“case contamination”), resistant to isolated gross errors at some cells \((i, j)\) (“cell contamination”), and applicable when some of the \(x_{ij}\) are missing (“irregularly spaced – or ‘incomplete’ – data”). Two approaches will be proposed for this problem. One deals with the three requirements stated above, and is based on ideas similar to MM-estimation [V. J. Yohai, Ann. Stat. 15, 642–656 (1987; Zbl 0624.62037)]. The other is a simple and fast estimator which can be applied to complete data with case- and cellwise contamination, and is based on applying a standard robust principal components estimate and smoothing the principal directions. Experiments with real and simulated data suggest that with complete data the simple estimator outperforms its competitors, while the MM estimator is competitive for incomplete data.

MSC:

62H25 Factor analysis and principal components; correspondence analysis
62F35 Robustness and adaptive procedures (parametric inference)

Citations:

Zbl 0624.62037
PDFBibTeX XMLCite
Full Text: DOI arXiv

References:

[1] Bali, JL; Boente, G.; Tyler, DE; Wang, J-L, Robust functional principal components: a projection-pursuit approach, Ann Stat, 39, 2852-2882 (2011) · Zbl 1246.62145 · doi:10.1214/11-AOS923
[2] Bay SD (1999) The UCI KDD Archive [http://kdd.ics.uci.edu], University of California, Irvine, Department of Information and Computer Science
[3] Boente, G.; Salibian-Barrera, M., S-estimators for functional principal component analysis, JASA, 110, 1100-1111 (2015) · Zbl 1373.62290 · doi:10.1080/01621459.2014.946991
[4] Cevallos Valdiviezo H (2016) On methods for prediction based on complex data with missing values and robust principal component analysis, PhD thesis, Ghent University (supervisors Van Aelst S. and Van den Poel, D.)
[5] Cleveland, WS, Robust locally weighted regression and smoothing scatterplots, JASA, 74, 829-836 (1979) · Zbl 0423.62029 · doi:10.1080/01621459.1979.10481038
[6] Gnanadesikan, R.; Kettenring, JR, Robust estimates, residuals, and outlier detection with multiresponse data, Biometrics, 28, 81-124 (1972) · doi:10.2307/2528963
[7] Grecki, T.; Krzyko, M.; Waszak, L.; Woyski, W., Selected statistical methods of data analysis for multivariate functional data, Stat Pap, 59, 153-182 (2018) · Zbl 1392.62173 · doi:10.1007/s00362-016-0757-8
[8] James, G.; Hastie, TG; Sugar, CA, Principal component models for sparse functional data, Biometrika, 87, 587-602 (2001) · Zbl 0962.62056 · doi:10.1093/biomet/87.3.587
[9] Lee, S.; Shin, H.; Billor, N., M-type smoothing spline estimators for principal functions, Comput Stat Data Anal, 66, 89-100 (2013) · Zbl 1471.62109 · doi:10.1016/j.csda.2013.03.022
[10] Locantore, N.; Marron, JS; Simpson, DG; Tripoli, N.; Zhang, JT; Cohen, KL, Robust principal components for functional data, Test, 8, 1-28 (1999) · Zbl 0980.62049 · doi:10.1007/BF02595862
[11] Maronna, R., Principal components and orthogonal regression based on robust scales, Technometrics, 47, 264-273 (2005) · doi:10.1198/004017005000000166
[12] Maronna, RA; Martin, RD; Yohai, VJ; Salibian-Barrera, M., Robust statistics: theory and methods (with R) (2019), Chichester: Wiley, Chichester · Zbl 1409.62009
[13] Rousseeuw, PJ; Croux, C., Alternatives to the median absolute deviation, JASA, 88, 1273-1283 (1993) · Zbl 0792.62025 · doi:10.1080/01621459.1993.10476408
[14] Yao, F.; Müller, H-G; Wang, J-L, Functional data analysis for sparse longitudinal data, JASA, 100, 577-590 (2005) · Zbl 1117.62451 · doi:10.1198/016214504000001745
[15] Yohai, VJ, High breakdown-point and high efficiency robust estimates for regression, Ann Stat, 15, 642-656 (1987) · Zbl 0624.62037 · doi:10.1214/aos/1176350366
[16] Yohai, VJ; Zamar, RH, High breakdown-point estimates of regression by means of the minimization of an efficient scale, J Am Stat Assoc, 83, 406-413 (1988) · Zbl 0648.62036 · doi:10.1080/01621459.1988.10478611
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.