×

Principal components for multivariate functional data. (English) Zbl 1464.62025

Summary: A principal component method for multivariate functional data is proposed. Data can be arranged in a matrix whose elements are functions so that for each individual a vector of \(p\) functions is observed. This set of \(p\) curves is reduced to a small number of transformed functions, retaining as much information as possible. The criterion to measure the information loss is the integrated variance. Under mild regular conditions, it is proved that if the original functions are smooth this property is inherited by the principal components. A numerical procedure to obtain the smooth principal components is proposed and the goodness of the dimension reduction is assessed by two new measures of the proportion of explained variability. The method performs as expected in various controlled simulated data sets and provides interesting conclusions when it is applied to real data sets.

MSC:

62-08 Computational methods for problems pertaining to statistics
62H25 Factor analysis and principal components; correspondence analysis
62R10 Functional data analysis

Software:

fda (R)
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] Acker, A.F., Absolute continuity of eigenvectors of time-varying operators, Proceedings of the American mathematical society, 42, 198-201, (1974) · Zbl 0281.47011
[2] Aneiros Pérez, G.; Cardot, H.; Estévez Pérez, G.; Vieu, P., Maximum ozone concentration forecasting by functional nonparametric approaches, Environmetrics, 15, 675-685, (2004)
[3] Aneiros Pérez, G.; Vieu, P., Semi-functional partial linear regression, Statistics & probability letters, 11, 1102-1110, (2006) · Zbl 1090.62036
[4] Boente, G.; Fraiman, R., Kernel-based functional principal components, Statistics & probability letters, 48, 335-345, (2000) · Zbl 0997.62024
[5] Coetzee, N.F., Nokes, W., Monismith, C., Metcalf, J., Mahoney, J., 2000. Full-scale/accelerated pavement testing: current status and future directions. Millennium Paper Series.; Coetzee, N.F., Nokes, W., Monismith, C., Metcalf, J., Mahoney, J., 2000. Full-scale/accelerated pavement testing: current status and future directions. Millennium Paper Series.
[6] Delicado, P., Another look at principal curves and surfaces, Journal of multivariate analysis, 77, 84-116, (2001) · Zbl 1033.62048
[7] Delicado, P., Dimensionality reduction when data are density functions, Computational statistics & data analysis, 55, 401-420, (2011) · Zbl 1247.62148
[8] Escabias, M.; Aguilera, A.M.; Valderrama, M.J., Functional PLS logit regression model, Computational statistics & data analysis, 51, 4891-4902, (2007) · Zbl 1162.62392
[9] Ferraty, F., Special issue: statistical methods and problems in infinite-dimensional spaces, Journal of multivariate analysis, 101, 305-490, (2010) · Zbl 1222.62083
[10] Ferraty, F.; Romain, Y., The Oxford handbook of functional data analysis, (2011), Oxford University Press
[11] Ferraty, F.; Vieu, P., Non parametric functional data analysis. theory and practice, (2006), Springer · Zbl 1119.62046
[12] Ferraty, F.; Vieu, P., Additive prediction and boosting for functional data, Computational statistics & data analysis, 53, 1400-1413, (2009) · Zbl 1452.62989
[13] Friedman, J.H.; Tukey, J.W., A projection pursuit algorithm for exploratory data analysis, IEEE transactions on computers, C-23, 881-890, (1974) · Zbl 0284.68079
[14] Golberg, M.A., The derivative of a determinant, American mathematical monthly, 79, 1124-1126, (1972) · Zbl 0287.34008
[15] Golub, G.H.; Van Loan, C.F., Matrix computations, (1996), Johns Hopkins University Press Baltimore · Zbl 0865.65009
[16] González-Manteiga, W.; Vieu, P., Introduction to the special issue on statistics for functional data, Computational statistics & data analysis, 51, 4788-4792, (2007) · Zbl 1162.62338
[17] Hastie, T.; Stuetzle, W., Principal curves, Journal of the American statistical association, 84, 502-516, (1989) · Zbl 0679.62048
[18] Hastie, T.; Tibshirani, R.; Friedman, J., ()
[19] Hyvärinen, A.; Oja, E., Independent component analysis: algorithms and applications, Neural networks, 13, 411-430, (2000)
[20] James, G.; Hastie, T.; Sugar, C., A principal component models for sparse functional data, Biometrika, 87, 587-602, (2000) · Zbl 0962.62056
[21] Lax, P.D., Linear algebra, (1997), Wiley New York · Zbl 0904.15001
[22] Li, P.L.; Chiou, J.M., Identifying cluster number for subspace projected functional data clustering, Computational statistics & data analysis, 55, 2090-2103, (2011) · Zbl 1328.62387
[23] Locantore, N.; Marron, J.S.; Simpson, D.G.; Tripoli, N.; Zhang, J.T.; Cohen, K.L., Robust principal component analysis for functional data, Test, 8, 1-73, (1999) · Zbl 0980.62049
[24] Manté, C.; Yao, A.-F.; Degiovanni, C., Principal component analysis of measures, with special emphasis on grain-size curves, Computational statistics & data analysis, 51, 4969-4983, (2007) · Zbl 1162.62375
[25] Müller, H.G., Functional modelling and classification of longitudinal data, Scandinavian journal of statistics, 23, 223-240, (2005) · Zbl 1089.62072
[26] Park, J.; Gasser, T.; Rousson, V., Structural components in functional data, Computational statistics & data analysis, 53, 3452-3465, (2009) · Zbl 1453.62171
[27] Preda, C.; Saporta, G., PLS regression on a stochastic process, Computational statistics & data analysis, 48, 149-158, (2005) · Zbl 1429.62224
[28] Ramsay, J.; Dalzell, C.J., Some tools for functional data analysis, Journal of the royal statistics society B, 53, 539-572, (1991) · Zbl 0800.62314
[29] Ramsay, J.; Silverman, B.W., Applied functional data analysis: methods and case studies, (2002), Springer New York · Zbl 1011.62002
[30] Ramsay, J.; Silverman, B.W., Functional data analysis, (2005), Springer New York · Zbl 1079.62006
[31] Rice, J.; Silverman, B.W., Estimating the Mean and the covariance structure nonparametrically when the data are curves, Journal of the royal statistics society B, 53, 233-243, (1991) · Zbl 0800.62214
[32] Sangalli, L.M., Secchi, P., Vantini, S., Vitelli, V., 2010. Joint clustering and alignment of functional data: an application to vascular geometries. MOX Report No. 09/2010. (http://mox.polimi.it/it/progetti/pubblicazioni/quaderni/09-2010.pdf; Sangalli, L.M., Secchi, P., Vantini, S., Vitelli, V., 2010. Joint clustering and alignment of functional data: an application to vascular geometries. MOX Report No. 09/2010. (http://mox.polimi.it/it/progetti/pubblicazioni/quaderni/09-2010.pdf · Zbl 1329.62289
[33] Schumaker, L., Spline functions: basic theory, (2007), Cambridge University Press Cambridge, Cambridge Mathematical Library · Zbl 1123.41008
[34] Shin, H., Partial functional linear regression, Journal of statistical planning and inference, 139, 3405-3418, (2009) · Zbl 1168.62358
[35] Shi, J.; Wang, B.; Murray-Smith, R.; Titterington, D., Gaussian process functional regresson modelling for batch data, Biometrics, 63, 714-723, (2007) · Zbl 1128.62053
[36] Silverman, B.W., Smoothed functional principal components analysis by the choice of norm, Annals of statistics, 24, 1-24, (1996) · Zbl 0853.62044
[37] Valderrama, M., Introduction to the special issue on modelling functional data in practice, Computational statistics, 22, 331-334, (2007)
[38] van der Linde, A, Variational Bayesian functional PCA, Computational statistics & data analysis, 53, 517-533, (2008) · Zbl 1452.62425
[39] Wang, S.; Jank, W.; Shmueli, G., Explaining and forecasting online auction prices and their dynamics using functional data analysis, Journal of business and economic statistics, 26, 144-160, (2008)
[40] Wold, H., Estimation of principal components and related models by iterative least squares, (), 391-420
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.