The Annals of Statistics

Weakly dependent functional data

Siegfried Hörmann and Piotr Kokoszka

Full-text: Open access

Abstract

Functional data often arise from measurements on fine time grids and are obtained by separating an almost continuous time record into natural consecutive intervals, for example, days. The functions thus obtained form a functional time series, and the central issue in the analysis of such data consists in taking into account the temporal dependence of these functional observations. Examples include daily curves of financial transaction data and daily patterns of geophysical and environmental data. For scalar and vector valued stochastic processes, a large number of dependence notions have been proposed, mostly involving mixing type distances between σ-algebras. In time series analysis, measures of dependence based on moments have proven most useful (autocovariances and cumulants). We introduce a moment-based notion of dependence for functional time series which involves m-dependence. We show that it is applicable to linear as well as nonlinear functional time series. Then we investigate the impact of dependence thus quantified on several important statistical procedures for functional data. We study the estimation of the functional principal components, the long-run covariance matrix, change point detection and the functional linear model. We explain when temporal dependence affects the results obtained for i.i.d. functional observations and when these results are robust to weak dependence.

Article information

Source
Ann. Statist., Volume 38, Number 3 (2010), 1845-1884.

Dates
First available in Project Euclid: 24 March 2010

Permanent link to this document
https://projecteuclid.org/euclid.aos/1269452656

Digital Object Identifier
doi:10.1214/09-AOS768

Mathematical Reviews number (MathSciNet)
MR2662361

Zentralblatt MATH identifier
1189.62141

Subjects
Primary: 62M10: Time series, auto-correlation, regression, etc. [See also 91B84]
Secondary: 60G10: Stationary processes 62G05: Estimation

Keywords
Asymptotics change points eigenfunctions functional principal components functional time series long-run variance weak dependence

Citation

Hörmann, Siegfried; Kokoszka, Piotr. Weakly dependent functional data. Ann. Statist. 38 (2010), no. 3, 1845--1884. doi:10.1214/09-AOS768. https://projecteuclid.org/euclid.aos/1269452656


Export citation

References

  • [1] Anderson, T. W. (1994). The Statistical Analysis of Time Series. Wiley, New York.
  • [2] Andrews, D. W. K. (1984). Nonstrong mixing autoregressive processes. J. Appl. Probab. 21 930–934.
  • [3] Andrews, D. W. K. (1991). Heteroskedasticity and autocorrelation consistent covariance matrix estimation. Econometrica 59 817–858.
  • [4] Antoniadis, A. and Sapatinas, T. (2003). Wavelet methods for continuous time prediction using Hilbert-valued autoregressive processes. J. Multivariate Anal. 87 133–158.
  • [5] Aue, A., Horváth, S., Hörmann L. and Reimherr, M. (2009). Detecting changes in the covariance structure of multivariate time series models. Ann. Statist. 37 4046–4087.
  • [6] Benko, M., Härdle, W. and Kneip, A. (2009). Common functional principal components. Ann. Statist. 37 1–34.
  • [7] Berkes, I., Gabrys, R., Horváth, L. and Kokoszka, P. (2009). Detecting changes in the mean of functional observations. J. R. Stat. Soc. Ser. B Stat. Methodol. 71 927–946.
  • [8] Berkes, I., Hörmann, S. and Schauer, J. (2009). Asymptotic results for the empirical process of stationary sequences. Stochastic Process. Appl. 119 1298–1324.
  • [9] Berkes, I., Horváth, L., Kokoszka, P. and Shao, Q.-M. (2005). Almost sure convergence of the Bartlett estimator. Period. Math. Hungar. 51 11–25.
  • [10] Berkes, I., Horváth, L., Kokoszka, P. and Shao, Q.-M. (2006). On discriminating between long-range dependence and changes in mean. Ann. Statist. 34 1140–1165.
  • [11] Besse, P., Cardot, H. and Stephenson, D. (2000). Autoregressive forecasting of some functional climatic variations. Scand. J. Statist. 27 673–687.
  • [12] Billingsley, P. (1968). Convergence of Probability Measures. Wiley, New York.
  • [13] Billingsley, P. (1999). Convergence of Probability Measures, 2nd ed. Wiley, New York.
  • [14] Bosq, D. (2000). Linear Processes in Function Spaces. Springer, New York.
  • [15] Bosq, D. and Blanke, D. (2007). Inference and Prediction in Large Dimensions. Wiley, Chichester.
  • [16] Bradley, R. C. (2007). Introduction to Strong Mixing Conditions 1, 2, 3. Kendrick Press, Heber City, UT.
  • [17] Cai, T. and Hall, P. (2006). Prediction in functional linear regression. Ann. Statist. 34 2159–2179.
  • [18] Cardot, H., Ferraty, F., Mas, A. and Sarda, P. (2003). Testing hypothesis in the functional linear model. Scand. J. Statist. 30 241–255.
  • [19] Cardot, H., Ferraty, F. and Sarda, P. (2003). Spline estimators for the functional linear model. Statist. Sinica 13 571–591.
  • [20] Chiou, J.-M. and Müller, H.-G. (2007). Diagnostics for functional regression via residual processes. Comput. Statist. Data Anal. 15 4849–4863.
  • [21] Chiou, J.-M., Müller, H.-G. and Wang, J.-L. (2004). Functional response models. Statist. Sinica 14 675–693.
  • [22] Cuevas, A., Febrero, M. and Fraiman, R. (2002). Linear functional regression: The case of fixed design and functional response. Canad. J. Statist. 30 285–300.
  • [23] Damon, J. and Guillas, S. (2002). The inclusion of exogenous variables in functional autoregressive ozone forecasting. Environmetrics 13 759–774.
  • [24] Dauxois, J., Pousse, A. and Romain, Y. (1982). Asymptotic theory for principal component analysis of a vector random function. J. Multivariate Anal. 12 136–154.
  • [25] Doukhan, P. (1994). Mixing: Properties and Examples. Springer, New York.
  • [26] Doukhan, P. and Louhichi, S. (1999). A new weak dependence and applications to moment inequalities. Stochastic Process. Appl. 84 313–343.
  • [27] Engle, R. F. (1982). Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation. Econometrica 50 987–1007.
  • [28] Giraitis, L., Kokoszka, P. S. and Leipus, R. (2000). Stationary ARCH models: Dependence structure and Central Limit Theorem. Econometric Theory 16 3–22.
  • [29] Giraitis, L., Kokoszka, P. S., Leipus, R. and Teyssière, G. (2003). Rescaled variance and related tests for long memory in volatility and levels. J. Econometrics 112 265–294.
  • [30] Gohberg, I., Golberg, S. and Kaashoek, M. A. (1990). Classes of Linear Operators, vol. 1. Operator Theory: Advances and Applications 49. Birkhaüser, Basel.
  • [31] Gordin, M. (1969). The central limit theorem for stationary processes. Dokl. Akad. Nauk SSSR 188 739–741.
  • [32] Granger, C. W. J. and Anderson, A. P. (1979). An Introduction to Bilinear Time Series Models. Vandenhoeck and Ruprecht, Göttingen.
  • [33] Hall, P. and Hosseini-Nasab, M. (2006). On properties of functional principal components. J. R. Stat. Soc. Ser. B Stat. Methodol. 68 109–126.
  • [34] Hamilton, J. D. (1994). Time Series Analysis. Princeton Univ. Press, Princeton, NJ.
  • [35] Hörmann, S. (2008). Augmented GARCH sequences: Dependence structure and asymptotics. Bernoulli 14 543–561.
  • [36] Hörmann, S. (2009). Berry-Esseen bounds for econometric time series. ALEA 6 377–397.
  • [37] Ibragimov, I. A. (1962). Some limit theorems for stationary processes. Teor. Veroyatn. Primen. 7 361–392.
  • [38] Kargin, V. and Onatski, A. (2008). Curve forecasting by functional autoregression. J. Multivariate Anal. 99 2508–2506.
  • [39] Kiefer, J. (1959). K-sample analogues of the Kolmogorov–Smirnov and Cramér-von Mises tests. Ann. Math. Statist. 30 420–447.
  • [40] Li, Y. and Hsing, T. (2007). On rates of convergence in functional linear regression. J. Multivariate Anal. 98 1782–1804.
  • [41] Malfait, N. and Ramsay, J. O. (2003). The historical functional model. Canad. J. Statist. 31 115–128.
  • [42] Mas, A. and Menneteau, L. (2003). Perturbation approach applied to the asymptotic study of random operators. Progr. Probab. 55 127–134.
  • [43] Maslova, I., Kokoszka, P., Sojka, J. and Zhu, L. (2009). Removal of nonconstant daily variation by means of wavelet and functional data analysis. Journal of Geophysical Research 114 A03202.
  • [44] Maslova, I., Kokoszka, P., Sojka, J. and Zhu, L. (2009). Estimation of Sq variation by means of multiresolution and principal component analyses. Journal of Atmospheric and Solar-Terrestrial Physics. To appear.
  • [45] Müller, H.-G. and Yao, F. (2008). Functional additive models. J. Amer. Statist. Assoc. 103 1534–1544.
  • [46] Müller, H.-G. and Stadtmüller, U. (2005). Generalized functional linear models. Ann. Statist. 33 774–805.
  • [47] Philipp, W. and Stout, W. F. (1975). Almost Sure Invariance Principles for Partial Sums of Weakly Dependent Random Variables. Mem. Amer. Math. Soc. 161. Amer. Math. Soc., Providence, RI.
  • [48] Pötscher, B. and Prucha, I. (1997). Dynamic Non-linear Econonometric Models. Asymptotic Theory. Springer, Berlin.
  • [49] Priestley, M. (1988). Nonlinear and Nonstationary Time Series Analysis. Academic Press, London.
  • [50] Ramsay, J. O. and Silverman, B. W. (2005). Functional Data Analysis. Springer, New York.
  • [51] Reiss, P. T. and Ogden, R. T. (2007). Functional principal component regression and functional partial least squares. J. Amer. Statist. Assoc. 102 984–996.
  • [52] Reiss, P. T. and Ogden, R. T. (2009). Smoothing parameter selection for a class of semiparametric linear models. J. R. Stat. Soc. Ser. B Stat. Methodol. 71 505–523.
  • [53] Reiss, P. T. and Ogden, R. T. (2010). Functional generalized linear models with images as predictors. Biometrics. To appear.
  • [54] Rosenblatt, M. (1956). A central limit theorem and a strong mixing condition. Proc. Natl. Acad. Sci. USA 42 43–47.
  • [55] Rosenblatt, M. (1959). Stationary processes as shifts of functions of independent random variables. J. Math. Mech. 8 665–681.
  • [56] Rosenblatt, M. (1961). Independence and dependence. In Proc. 4th Berkeley Sympos. Math. Statist. Probab. 2 431–443. Univ. California Press, Berkeley, CA.
  • [57] Rosenblatt, M. (1971). Markov Processes. Structure and Asymptotic Behavior. Springer, New York.
  • [58] Stadtlober, E., Hörmann, S. and Pfeiler, B. (2008). Qualiy and performance of a PM10 daily forecasting model. Athmospheric Environment 42 1098–1109.
  • [59] Stine, R. (1997). Nonlinear time series. In Encyclopedia of Statistical Sciences (S. Kotz, C. Y. Read and D. L. Banks, eds.) 430–437. Wiley, New York.
  • [60] Tong, H. (1990). Non-linear Time Series: A Dynamical System Approach. Oxford Univ. Press, New York.
  • [61] Vostrikova, L. J. (1981). Detection of “disorder” in multidimensional random processes. Sov. Dokl. Math. 24 55–59.
  • [62] Wu, W. (2005). Nonlinear system theory: Another look at dependence. Proc. Natl. Acad. Sci. USA 102 14150–14154.
  • [63] Wu, W. (2007). Strong invariance principles for dependent random variables. Ann. Probab. 35 2294–2320.
  • [64] Yao, F. and Lee, T. (2006). Penalized spline models for functional principal component analysis. J. R. Stat. Soc. Ser. B Stat. Methodol. 68 3–25.
  • [65] Yao, F., Müller, H.-G. and Wang., J.-L. (2005). Functional linear regression analysis for longitudinal data. Ann. Statist. 33 2873–2903.
  • [66] Yao, F., Müller, H.-G. and Wang., J.-L. (2005). Functional data analysis for sparse longitudinal data. J. Amer. Statist. Assoc. 100 577–590.