Abstract
This paper shows how the innovation approach developed by Wiener (1949), Kaiman (1960) and Box and Jenkins (1970) has found wide application in modern nonlinear time series analysis. Nonlinear models, such as the chaos, stochastic or deterministic differential equation models, neural network models and nonlinear AR models developed in the last two decades are reviewed as useful causal models in time series analysis for nonlinear dynamic phenomena in many scientific fields. The merit of the use of the innovation approach in conjunction with these new models is pointed out. Further, the computational efficiency and advantage of RBF-AR models over RBF neural network models is demonstrated in real data analysis of EEG time series of subjects with epilepsy. The advantage of multivariate RBF-ARX models in the modeling of thermal power plants is also shown using numerical results.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Akaike H. (1969), “Fitting autoregressive models for prediction,” Ann. Inst. Stat. Math., 21, 243–247.
Biscay R., J.C. Jimenez, J.J. Riera, and P.A. Valdes Sosa (1996), “Local linearization method for the numerical solution of stochastic differential equations,” Ann. Inst. Stat. Math., 48, 631–644.
Bouchaud J.P. and R. Cont (1998), “A Langevin approach to stock market fluctuations and crashes,” Eur. Phys. J., B6, pp. 543–550.
Box G.E.P. and G.M. Jenkins (1970), Time Series Analysis: Forecasting and Control, Holden-Day, San Francisco.
Campbell M.J. and A.M. Walker (1977), “A survey of statistical work on the McKenzie river series of annual Canadian lynx trappings for the year 1821–1934, and a new analysis,” J. Roy. Statist. Soc., A 140, 411–431.
Cichocki A. and R. Unbehauen (1993), “Neural Networks for Optimization and Signal Processing,” J. Wiley and Sons, Chichester/New York/Brisbane/Toronto/Singapore.
Doob J.L. (1953), Stochastic Processes, John Wiley and Sons Inc.
Dyn N. (1989a), “Interpolation and approximation by radial and related functions,” in Approximation theory, VI, Academic Press, London, pp. 211–234.
Dyn N. (1989b), “Interpolation by piecewise-linear radial basis functions,” J. Approximation Theory, 59, 202–223.
Frost P.A. and T. Kailath (1971), “An innovation approach to least squares estimation Part III: nonlinear estimation in white noise,” IEEE Trans on Aut. Control, AC-13, pp. 646–655.
Haber R. and L. Keviczky (1999), Nonlinear system identification — input-output modeling approach, Volume 1: Nonlinear system structure identification. Kluwer Academic Publishers, The Netherlands.
Iino M. and T. Ozaki (2000), “A Nonlinear Model for Financial Dynamics,” in Proceeding of the International Symposium on Frontiers of Time Series Modeling, ISM Tokyo Japan.
Ito K. (1980), “Chaos in the Rikitake two-disc dynamo system,” Earth Planet. Sci. Lett., 51, 451–456.
Jimenez J.C., I. Shoji, and T. Ozaki (1999), “Simulation of stochastic differential equations through the local linearization method: a comparative study,” J. Stat. Phys., 94, 587–602.
Jimenez J.C. and T. Ozaki (2000a), “Linear estimation of continuous-discrete linear state space models with multiplicative noise,” Res. Memo., 772, The Institute of Statistical Mathematics, Tokyo.
Jimenez J.C. and T. Ozaki (2000b), “Local Linearization filters for nonlinear continuous-discrete state space models with multiplicative noise,” Res. Memo., 773, The Institute of Statistical Mathematics, Tokyo.
Kailath T. (1968), “An innovation approach to least squares estimation Part I: linear filtering in additive white noise,” IEEE Trans on Aut. Control, AC-13, pp. 646–655.
Kalman R.E. (1960), A New Approach to Linear Filtering and Prediction Problems, Trans. ASME, Series D, J. of Basic Eng., 82, 35–45.
Kloeden P.E. and E. Platen (1995), Numerical Solution of Stochastic Differential Equations, New York, Springer.
Kolmogorov A.N. (1931), “On analytical methods in probability theory,” Math. Ann., 104, 415–458.
Kolmogorov A.N. (1941), “Interpolation and extrapolation of stationary random sequences,” Bull. Acad. Sci. USSR, Math. Ser., 5, 3–14.
Lapedes A. and R. Farber, (1987), “Neural Networks work,” Neural Information Systems, Academic Institute of Physics, pp. 442–456.
Lopes da Silva F.H., A. van Rotterdam, P. Barts, E. van Heusden, and W. Burr (1976), “Models of neuronal populations: the basic mechanisms.” In: M.A. Corner and D.F. Swaab (Eds.), Perspective in Brain Research, Prog. Brain Res., 45, 281–308.
Masani P. and N. Wiener (1959), “Nonlinear prediction,” in Probability and Statistics, John Wiley and Sons, pp. 190–212.
May R. (1976), “Simple mathematical models with very complicated dynamics,” Nature, 261, 459–467.
Mil’stein G.N. (1995), Numerical Integration of Stochastic Differential Equations, Dordrecht: Kluwer Academic Publishers.
Ozaki T. (1980), “Nonlinear time series models for nonlinear random vibrations,” J. of Appl Prob., 17, 84–93.
Ozaki T. (1985a), “Nonlinear time series models and dynamical systems,” in Handbook of Statistics, 5, E.J. Hannan et al. (Eds.), North-Holland.
Ozaki T. (1985b), “Statistical identification of storage models with application to stochastic hydrology,” Water Resources Bulletin, 21.
Ozaki T. (1986), “Local Gaussian modelling of stochastic dynamical systems in the analysis of nonlinear random vibrations,” in Essays in Time Series and Allied Processes, Festschrift in honour of Prof. E.J. Hannan, Probability Trust.
Ozaki T. (1992), “Identification of nonlinearities and non-Gaussianities in time series,” in New Directions in Time Series Analysis, Part I, D.R. Brillinger et al. (Eds.), IMA Volumes in Mathematics and Its Application, Springer Verlag, 45, 227–264.
Ozaki T. (1993a), “A local linearization approach to nonlinear filtering,” Int. J. Control, 57, 75–96.
Ozaki T. (1993b), “Non-Gaussian characteristics of Exponential Autoregressive processes,” in Developments in Time Series Analysis, Festschrift in honour of Prof. Maurice Priestley, T. Subba Rao (Ed.), Chapman and Hall, pp. 257–273.
Ozaki T. and H. Oda (1978), “Nonlinear time series model identification by Akaike’s information criterion,” Information and Systems, B. Dubuisson (Ed.), Pergamon Press, pp. 83–91.
Ozaki T. and V.H. Ozaki (1989), “Statistical identification of nonlinear dynamics in macroeconomics using nonlinear time series models,” in Statistical Analysis and Forecasting of Economic Structural Change, P. Hackl (Ed.), Springer-Verlag, pp. 345–365.
Ozaki T., P.A. Valdes Sosa, and V. Haggan-Ozaki (1997), “Reconstructing Nonlinear Dynamics from Time Series: with Application to Epilepsy Data Analysis,” J. Signal Processing, 3, 153–162.
Ozaki T., J.C. Jimenez, and V. Haggan-Ozaki (2000), “Role of the likelihood function in the estimation of chaos models,” J. Time Series Analysis, 21, 363–387.
Ozaki T., J.C. Jimenez, M. Iino, Z. Shi, and S. Sugawara (2001), “Use of stochastic differential equation models in financial time series analysis: Monitoring and control of currencies in exchange market,” Proceeding of 3rd Japan-US joint seminar on statistical time series analysis, Kyoto, Japan, June 18–22, Sponsored by JSPS-NSF, June 2001, pp. 17–24.
Park J. and I.W. Sandberg (1991), “Universal approximation using radial basis function networks,” Neural Comput., 3, 247–257.
Park J. and I.W. Sandberg (1993), “Approximation and radial basis networks,” Neural Comput., 5, 305–306.
Peng. H., T. Ozaki, and Y. Toyoda (2001), “Nonlinear system identification using radial basis function-based signal-dependent ARX model,” Proc. of 5th IFAC Symposium on Nonlinear Control Systems, Saint-Petersburg, Russia, July 4–6, 2001.
Pogio T. and F. Girosi (1990), “Networks for Approximation and Learning,” Proceeding of the IEEE, 78, 1461–1497.
Roberts S. and L. Tarassenko (1995), “Automated sleep EEG analysis using an RBF network,” in Applications of Neural Networks, A. Murray (Ed.), Kluwer Academic Publishers, pp. 305–322.
Rikitake T (1958), “Oscillations of a system of disc dynamos,” Proc. Camb. Philos. Soc, 54, 89–105.
Shi Z., Y. Tamura, and T. Ozaki (1999), “Nonlinear time series modelling with radial basis function-based state-dependent autoregressive model,” Intern. J. of System science, 30, 717–727.
Shoji I. and T. Ozaki (1997), “Comparative study of estimation methods for continuous time stochastic processes,” Journal of Time Series Analysis, 18, 485–506.
Shoji I. and T. Ozaki (1998), “A statistical method of estimation and simulation for systems of stochastic differential equations,” Biometrika, 85, 240–243.
Shoji I. (1998), “A comparative study of maximum likelihood estimators for nonlinear dynamical system models,” Int. J. Control, 71, 391–404.
Tong H. (1977), “Some comments on the Canadian lynx data,” J. Roy. Soc, A 140, 432–436.
Toyoda Y., K. Oda, and T. Ozaki (1997), “The nonlinear system identification method for advanced control of the fossil power plants”. Proceedings of 11 th IFAC Symposium on System Identification, Fukuoka, Japan, pp. 8–11.
Valdes Sosa P.A., J.C. Jimenez, J. Riera, R. Biscay, and T. Ozaki (1999a), “Nonlinear EEG analysis based on a neural mass model,” Biol. Cybern., 81, 415–424.
Valdes Sosa P., J. Bosch, s, N. Trujillo, R. Biscay, F. Morales, J.L. Hernandez, and T. Ozaki (1999b), “The Statistical Identification of Nonlinear Brain Dynamics: A Progress Report,” in “Nonlinear Dynamics and Brain Functioning,” Nova Science Publishers, edited by N. Pradhan, P.E. Rapp, and R. Sreenivasan.
Vesin J. (1993), “An amplitude-dependent autoregressive signal model based on a radial basis function expansion,” In Proc. Intern. Conf. Acoust. Speech Signal Process., Minnesota, 3, 129–132.
Wold H. (1938), A Study in the Analysis of Stationary Time Series, Uppsala, Sweden: Almqvist and Wicksell.
Wiener N. (1949), Extrapolation, Interpolation, and Smoothing of Stationary Time Series with Engineering Applications, New York: Wiley. Originally issued in MIT Radiation Lab. Report, Cambridge, Mass., February 1942.
Wiener N. (1961), Cybernetics, 2nd edition, MIT Press, Cambridge, Mass.
Wong E. (1963), “The construction of a class of stationary Markoff process,” Proc. Amer. Math. Soc. Symp. Appl. Math., 16, 264–276.
Zetterberg L.H., L. Kristiansson, and K. Mossberg (1978), “Performance of a model for a local neural populations,” Biol. Cybern., 31, 15–26.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag New York, LLC
About this paper
Cite this paper
Ozaki, T., Jimenez, J.C., Peng, H., Ozaki, V.H. (2004). The Innovation Approach to the Identification of Nonlinear Causal Models in Time Series Analysis. In: Brillinger, D.R., Robinson, E.A., Schoenberg, F.P. (eds) Time Series Analysis and Applications to Geophysical Systems. The IMA Volumes in Mathematics and its Applications, vol 139. Springer, New York, NY. https://doi.org/10.1007/978-1-4684-9386-3_11
Download citation
DOI: https://doi.org/10.1007/978-1-4684-9386-3_11
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-1971-7
Online ISBN: 978-1-4684-9386-3
eBook Packages: Springer Book Archive