Abstract
In this paper the problem of functional filtering of an autoregressive Hilbertian (ARH) process, affected by additive Hilbertian noise, is addressed when the functional parameters defining the ARH(p) equation are unknown. The maximum-likelihood estimation of such parameters is obtained from the implementation of an expectation-maximization algorithm. Specifically, a finite-dimensional matrix approximation of the state equation is considered from its diagonalization in terms of the spectral decomposition of the functional parameters involved (Principal-Oscillation-Pattern-based diagonalization). The Expectation step and maximization step are then computed from the forward Kalman filtering followed by a backward Kalman smoothing recursion in terms of the Fourier coefficients associated with such a decomposition.
Similar content being viewed by others
References
Alter O (2000) Singular value decomposition for genome-wide expression data processing and modelling. Proc Natl Acad Sci 97:10101–10106 (Online)
Alter O, Golub GH (2003) Generalized singular value decomposition for comparative analysis of genome-scale expression data sets of two different organisms. Proc Natl Acad Sci 100:3351–3356 (Online)
Antoniadis A, Sapatinas T (2003) Wavelet methods for continuous-time prediction using Hilbert-valued autoregressive processes. J Multivar Anal 87:133–158
Bar-Joseph Z, Gerber G, Jaakkola T, Gifford D, Simon I (2003) Continuous representation of time-series gene expression data. J Comput Biol 10:341–356
Bar-Joseph Z (2004) Analyzing time series gene expression data. Bioinformatics 20:2493–503
Besse P, Cardot H, Stephenson DB (2000) Autoregressive forecasting of some functional climatic variations. Scand J Stat 27:673–687
Bosq D (1991) Nonparametric statistics for stochastic processes, estimation and prediction. Lectures Notes in Statistics, vol 110. Springer, New York
Bosq D (2000) Linear processes in function spaces. Springer, New York
Bosq D (2008) A note on asymptotic parametric prediction. J Stat Plann Infer. doi:10.1016/j.jspi.2008.07.018
Bosq D, Blanke D (2007) Inference and prediction in large dimensions, Wiley Series in Probability and Statistics. Wiley, New York
Cardot H (1997) Contribution á l’estimation et á la prévision statistique de doneées fonctionnelles. Ph.D. thesis, University of Toulouse 3, France
Cavanaugh JE (1997) Unifying the derivation for the Akaike and corrected Akaike criteria. Stat Prob Lett 33:202–208
Christakos G (1985) Recursive parameter estimation with applications in Earth sciences. Math Geol 17:489–515
Christakos G (1988) On-line estimation of nonlinear physical systems. Math Geol 20:111–133
Christakos G (2005) Random field models in earth sciences. Academic Press, San Diego
Christakos G, Hristopulos DT (1998) Spatiotemporal environmental health modelling. Kluwer, Boston
Damon J, Guillas S (2005) Estimation and simulation of autoregressive hilbertian processes with exogenous variables. Stat Infer Stoch Proc 8:185–204
Dautray R, Lions JL (1992) Mathematical analysis and numerical methods for science and technology 3. Spectral theory and applications. Springer, Berlin
Dunford N, Schwartz JT (1971) Linear operators, part iii, spectral operators. Wiley, New York Interscience
Ferraty F, Vieu P (2006) Nonparameric functional data analysis, Springer series in statistics. Springer, New York
Germain F, Doisy A, Ronot X, Tracqui P (1999) Characterization of cell deformation and migration using a parametric estimation of image motion. IEEE Trans Biomed Eng 46:584–600
Goia A (2003) Selection model in functional linear regression models for scalar response. In: Ferligoj A, Mrvar A (eds) Developments in Applied Statistics, Metodoloski zvezki, vol 19, FDV, Ljubljana
Guillas S (2001) Rates of convergence of autocorrelation estimates for autoregressive Hilbertian processes. Stat Prob Lett 55:281–291
Hall P, Poskitt DS, Presnell B (2001) A functional data-analytic approach to signal discrimination. Technometrics 43:1–9
Haoudi A, Bensmail H (2006) Bioinformatics and data mining in proteomics. Expert Rev Proteomics 3:333–343
Kato T (1995) Perturbation theory of linear operators. Springer, New York
Klevezc RR, Murray DB (2001) Genome wide oscillations in expression: wavelet analysis of time series data from yest expression arrays uncovers the dynamic architecture of phenotype. Mol Biol Rep 28:73–82
Leng X, Müller H-G (2006) Classification using functional data analysis for temporal gene expression data. Bioinformatics 22:68–76
Mas A (1999) Normalité asymptotique de l’estimateur empirique de l’opérateur d’autocorrélation d’un processus ARH(1). C R Acad Sci Paris, 329 Série I, pp 899–902
Mas A (2007) Weak convergence in the functional autoregressive model. J Multivar Anal 98:1231–1261
Matheron G, Traité de géoestatistique apliquée, tome I, mémoires du Bureau de Recherches Géologiques et Miniéres, Editions Bureau de Recherches Geologiques et Miniéres, Paris, 24, 1962
Merlévede F (1996) Processus linéaires Hilbertiens: inversibilité, theorémes limites, estimation et prévision, Ph.D. Thesis, University of París 6, France
Monk NAM (2003) Unravelling nature’s networks. Biochem Soc Trans 31:1457–1561
Mourid T (1995) Contribution á la statistique des processus autiregréssifs á temps continu. Ph.D. thesis, University of Paris 6, France
Mourid T, Bensmain N (2006) Sieves estimator of the operator of a functional autoregressive process. Stat Prob Lett 76:93–108
Müller HG (2005) Functional modelling and classification of longitudinal data. Scand J Statist 32:223–240
Müller HG, Stadmüller U (2005) Generalized functional models. Ann Stat 33:774–885
Pumo B (1992) Estimation et prévision de processus autirégressifs fonctionnels. Application aux processus á temps continu. Ph.D. thesis, University of París 6, France
Ramsay JO, Silverman BW (2005) Functional data analysis, Springer series in statistics. Springer, New York
Raychadhuri S, Stuart JM, Altman RB (2000) Principal component analysis to summarize microarray experiments: application to sporulation time series. Pacif Sympos Bicomp 5:452–63
Ruiz-Medina MD, Salmerón R, Angulo JM (2007) Kalman filtering from POP-based diagonalization of ARH(1). Comput Stat Data Anal 51:4994–5008
Salmerón R, Ruiz-Medina MD (2009a) Multispectral decomposition of FAR(p) models. Stoch Env Res Risk Assess 23:289–297
Salmerón R, Ruiz-Medina MD (2009b) Functional SEM algorithm in ARH(1) models. XXXI Congreso Nacional de Estadística e Investigación Operativa, Spain
Schwartz G (1978) Estimating the dimension of a model. Ann Stat 6:461–464
Shibata R (1981) An optimal selection of regression variables. Biometrika 68:45–54
Song JJ, Lee H-J, Morris JS, Kangd S (2007) Clustering of time-course gene expression data using fuctional data analysis. Comput Biol Chem 31:265–274
Vieu P (1995) Order choice in nonlinear autoregressive models. Statistics 26:307–328
Yao F, Müller HG, Clifford AJ, Dueker SR, Lin Follett J, Buchholz BAY, Vogel JS (2003) Shrinkage estimation for functional principal component scores with application to the population kinetics of plasma folate. Biometrics 59:676–685
Yao F, Müller HG, Wang JL (2005) Functional data analysis for sparse longitudinal data. J Am Stat Assoc100:577–590
Acknowledgments
This work has been supported in part by projects MTM2008-03903, MTM2005-08597 of the DGI, MEC, and P05-FQM-00990, P06-FQM-02271 of the Andalousian CICE, Spain.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ruiz-Medina, M.D., Salmerón, R. Functional maximum-likelihood estimation of ARH(p) models. Stoch Environ Res Risk Assess 24, 131–146 (2010). https://doi.org/10.1007/s00477-009-0306-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-009-0306-2