Abstract
The purpose of this expository paper is to give a survey of rigorous asymptotic results in the theory of stochastic complexity. We provide a brief survey of the early developments and the ideas which culminated in a basic inequality on universal codes. The rest of the paper is devoted to results showing that the lower bound given by the basic inequality is achieved in many important cases. The asymptotic behaviour of three kinds of predictive stochastic complexities associated with ARMA processes will be established. Finally, extensions of the ARMA results for multivariable systems, for Ljung’s scheme, and for nonparametric problems will be considered.
On leave from Computer and Automation Institute of the Hungarian Academy of Sciences, Budapest. The work of the first author was supported by NSERC, Canada.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
H. Akaike, Statistical predictor identification Ann. Inst. Statis. Math., 22 (1970), pp.202–217.
H. Akaike, A new look at the statistical model identification IEEE Trans. Aut. Cont., AC-19 (1974), pp.716–723.
K.J. Åström, and B. Wittenmark, On self-tuning regulators Automatica, 9 (1973), pp.185–199.
J. Baikovicius and L. Gerencsér, Change point detection in a stochastic complexity framework To appear in the Proc of the 29-th IEEE CDC (1990).
A.R. Barron and T.M. Cover, Minimum complexity density estimation Submitted to IEEE Trans. Inf. Th. (1989).
A.R. Barron and C. Sheu, Approximation of densities by sequences of exponential families Submitted to the Annals of Statistics (1988).
R.J. Bhansali, Effects of not knowing the order of an autoregressive process on the squared error of prediction-I J. Amer. Statist. Assoc, 76 (1981), pp.588–597.
P.E. Caines, Linear Stochastic Systems Wiley, 1988.
P.E. Caines and M. Gürsoy, On the L ∞ consistency of L 2 estimators Systems and Control Letters, 12 (1989), pp.71–76.
L.D. Davisson, Prediction error of stationary Gaussian time series of unknown covariance IEEE Trans. Inform. Theory, IT-19 (1965), pp.783–795.
A.P. Dawid, Present Position and Potential Developments: Some Personal Views, Statistical Theory, The Prequential Approach J.R.S.S. A, Part 2, 147 (1984), pp.278–292.
M. Deistler and B.D.O. Anderson, Linear dynamic errors-in-variable models. Some structure theory J. of Econometrics, 41 (1989), pp.39–63.
L. Gerencsér, On a class of mixing processes Stochastics, 26 (1989a), pp.165–191.
L. Gerencsér, Strong approximation of the recursive prediction error estimator of the parameters of an ARMA process McGill Research Centre for Intelligent Machines, TR-CIM 89–8, Submitted to Systems and Control Letters (1991).
L. Gerencsér, On Rissanen’s predictive stochastic complexity for stationary ARMA processes McGill Research Centre for Intelligent Machines. TR-CIM-89–5. Submitted to Statistical Planning and Inference (1992).
L. Gerencsér, Some new results in the theory of recursive identification Proc. of the 28-th IEEE Conference on Decision and Control, Tampa, Florida, Vol.1, (1989), pp. 242–248.
L. Gerencsér, Asymptotics of Risannen’s predictive stochastic complexity: from parametric to nonparametric models Proc. of the 28-th IEEE CDC, Tampa, Florida, Vol. 2 (1989), pp.1210–1216.
L. Gerencsér, Strong approximation of the maximum likelihood estimator of the parameters of an ARMA process Proc. of the 28-th IEEE CDC, Tampa, Florida, Vol. 1 (1989), pp.630–631.
L. Gerencsér, Almost sure asymptotics of Rissanen’s predictive stochastic complexity In Realization and Modelling in System Theory. Proc. of the International Symposium on the MTNS-89, (eds. M.A. Kaashoek, J.H. van Schuppen and A.C.M. Ran), Vol. I (1990, pp. 429–436, Bikhauser, Boston.
L. Gerencsér, On the martingale approximation of the estimation error of ARMA parameters Systems and Control Letters, 15 (1990), pp. 417–423.
L. GERENCSÉR, Rate of convergence of recursive estimators To appear in the SIAM J. Control and Opt. (1992).
L. Gerencsér, Stochastic complexity in identification and adaptive control Proc. of the 11-th IFAC World Congress, Tallin, USSR, Vol. II (1990).
L. Gerencsér, Closed loop parameter identifiability and adaptive control of a linear stochastic system Systems and Control Letters, 15 (1990) pp. 411–416.
L. Gerencsér, Strong approximation of vector-valued stochastic integrals Statistics and Probability Letters, 12 (1991), pp. 201–207.
L. Gerencsér, Fixed gain estimation of ARMA parameters Submitted to the IEEE Trans, on Automatic Control, (1991).
L. Gerencsér, Predictive stochastic complexity associated with fixed gain estimators Submitted to IEEE Trans. Automatic Control, (1991).
L. Gerencsér and J. Rissanen, A prediction bound for Gaussian ARMA processes Proc. of the 25-th IEEE Conference on Decision and Control, Athens, Vol. 3 (1986), pp.1487–1490.
L. Gerencsér and Zs. Vágó, Model selection in continuous time Proc. of the 30-th IEEE Conference on Decision and Control, Vol. 2 (1991), pp. 959–962.
P. Hall and E.J. Hannan, On stochastic complexity and nonparametric density estimation Biometrika, 75 (1988), pp.705–719.
E.J. Hannan, and M. Deistler, The statistical theory of linear systems Wiley (1988).
E.J. Hannan, A.J. McDougall and D.S. Poskitt, Recursive estimation of autoregressions J. Roy, Stat. Soc. Ser B. 51 (1989), To appear.
E.J. Hannan and B.G. Quinn, The determination of the order of an autoregression J. Royal Statist. Soc. B., 41 (1979), pp.190–195.
E.J. Hannan and J. Rissanen, Recursive estimation of ARMA order Biometrika 69, 1 (1982), pp.81–94.
E.M. Hemerley and M.H.A. Davis, Strong consistency of the PLS criterion for order determination of autoregressive processes Ann. Stat., 17 (1989), pp.941–946.
A. Heunis, Rates of convergence for an adaptive filtering algorithm driven by stationary dependent data Submitted to SIAM J. Control and Opt. (1991).
P. Kabaila, On Rissanen’s lower bound on the accumulated prediction error J. of Time Series Analysis (1988).
L. Kavalieris, The estimation of the order of an autoregression using recursive residuals and cross-validation J. of Time Series analysis, 10 (1989), pp.271–281.
A.N. Kolmogorov and V.A. Upsenskii, Algorithms and randomness Theoria Veroyatnostey i ee Primeneniya (Theory of Probabilities and its Applications), 32–33 (1987), pp.425–455.
Yu. A. Kutoyants, Estimation of stochastic processes Publishing House of the Armenian Academy of Sciences, Yerevan (in Russian, 1980) and Heldermann, Berlin (1989).
T. Lai, Private communication.
A. Lempel AND J. Ziv, On the complexity of finite sequence IEEE Trans. Inf. Theory, IT-22 (1974), pp.75–81.
L. Levin, On the notion of a random sequence Soviet Math. Doklady 14 (1973), pp. 1413–1416.
L. Ljung, Analysis of recursive stochastic algorithms IEEE Trans, on Automatic Control, (1977), AC-22, pp.551–575.
L. Ljung and P. Caines, Asymptotic normality of prediction error estimation for approximate system models Stochastics, 31 (1979), pp.29–46.
L. Ljung and T. Söderström, Theory and practice of recursive identification MIT Press (1984).
J.M. Maciejowski, Model discrimination using an algorithmic information criterion Automatica, 15 (1979), pp.579–593.
P. Martin-Lov, The notion of redundancy and its use as a quantitative measure of the discrepancy between a statistical hypothesis and a set of observational data Sc. J. Statistics, 1 (1974), pp.3–18.
E. Parzen, Time series, statistics and information IMA Preprint #663, (1990).
J.R. Quinlan and R.L. Rivest, Inferring decision trees using minimum description length principle Information and Computation, 80 (1989), pp.227–248.
J. Rissanen, Modelling by shortest data description Automatica, 14 (1978), pp.465–471.
J. Rissanen, Order estimation by accumulated prediction errors Essays in Time Series and Allied Processes (eds. J. Gani, M.B. Priestley), pp.55–61 (1984).
J. Rissanen, Universal coding, information, prediction, and estimation IEEE Trans. Inf. Theory, Vol. IT-30 (1984), pp.629–636.
J. Rissanen, Stochastic complexity and modeling Annals of Statistics 14 (1986), pp.1080–1100.
J. Rissanen, A predictive least squares principle IMA J. Math. Control Inform., 3 (1986b), pp.211–222.
J. Rissanen, Stochastic complexity (with discussions) J. of the Royal Statistical Society, Series B, 49 (1987), No 3.
J. Rissanen, Stochastic complexity in statistical inquiry World Scientific Publisher (1989).
J. Rissanen, Proc. of 27th IEEE CDC, 3 (1988).
J. Rissanen, T.P. Speed and B. Yu, Density estimation by stochastic complexity IEEE Trans, on Information Theory, 38 (1992), pp. 315–323.
G. Schwarz, Estimating the dimension of a model Annals of Statistics 6 (1978), pp.461–464.
R. Shibata, Asymptotically efficient selection of the order of the model for estimating parameters of a linear process The Ann. Stat., 8 (1980), pp.147–164.
R. Shibata, Selection of the number of regression variables; A mini-max choice of generalized FPE’ Annals of Statistics, 38 (1980), pp.459–474.
V. Solo, On the convergence of AML IEEE Trans. Aut. Cont., AC-24 (1979), pp.958–962.
T. Söderström and P. Stoica, System identification (1989), Prentice Hall.
Zs. Vágó and L. Gerencsér, Almost sure asymptotics of stochastic complexity for i.i.d. sequences Submitted for publication.
G. Yin, Recent progress in parallel stochastic approximations In: Topics in Stochastic Systems: Modelling, Identification and Adaptive Control (eds. L. Gerencser and P.E. Caines), Lecture Notes in Control and Information Sciences, 161, (1991), pp. 159–184.
C.S. Wallace and D.M. Boulton, An information measure for classification Computing Journal, 11 (1968), pp.185–195.
M. Wax, Order selection for AR models by predictive least squares IEEE Trans., Acoustics, Speech, and Signal Processing, 36, (1988), pp.581–588.
C.Z. Wei, On predictive least squares principles Submitted for publication, (1990).
A.K. Zhvonkin and L.A. Levin, The complexity of finite objects and the development of the concepts of information and randomness by means of the theory of algorithms Russian Mathematical Surveys, 25 (1970), pp. 83–124.
J. Ziv and A. Lempel, Compression of individual sequences via variable-rate encoding IEEE Trans. Inform. Theory, IT-24 (1978), pp.530–536.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1993 Springer-Verlag New York, Inc.
About this paper
Cite this paper
Gerencsér, L., Rissanen, J. (1993). Asymptotics of Predictive Stochastic Complexity. In: Brillinger, D., Caines, P., Geweke, J., Parzen, E., Rosenblatt, M., Taqqu, M.S. (eds) New Directions in Time Series Analysis. The IMA Volumes in Mathematics and its Applications, vol 46. Springer, New York, NY. https://doi.org/10.1007/978-1-4613-9296-5_6
Download citation
DOI: https://doi.org/10.1007/978-1-4613-9296-5_6
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4613-9298-9
Online ISBN: 978-1-4613-9296-5
eBook Packages: Springer Book Archive