Abstract
Following a Markov chain approach, this paper establishes asymptotic properties of the least squares estimator in nonlinear autoregressive (NAR) models. Based on conditions ensuring the stability of the model and allowing the use of a strong law of large number for a wide class of functions, our approach improves some known results on strong consistency and asymptotic normality of the estimator. The exact convergence rate is established by a law of the iterated logarithm. Based on this law and a generalized Akaike's information criterion, we build a strongly consistent procedure for selection of NAR models. Detailed results are given for familiar nonlinear AR models like exponential AR models, threshold models or multilayer feedforward perceptions.
Similar content being viewed by others
References
Akaike, H. (1969). Fitting autoregressive models for prediction, Ann. Inst. Statist. Math., 21, 243–247.
Attali, J. (1998). Chaines de Markov Stables, PhD thesis, SAMOS, Université Paris I.
Borovkov, A. A. (1991). Lyapounov functions and ergodicity of multidimensional Markov chains, Theory Probab. Appl., 36(1), 1–18.
Borovkov, A. A. and Korshunov, D. (1994). Ergodicity in a sense of weak convergence, equilibrium-type identities and large deviations for Markov chains, Probability Theory and Mathematical Statistics: Proceedings of the Sixth Vilnius Conference (1993) (eds. B. Grigelionis, J. Kubilius, H. Pragauskas and V. Statulevicius), 89–98, TEV, Vilnius.
Cottrell, M., Girard, B., Girard, Y., Mangeas, M. and Muller, C. (1995). Neural modeling for time series: a statistical stepwise method for weight elimination, I.E.E.E. Trans. Neural Networks, 6, 1355–1364.
Doukhan, P. and Ghindès, M. (1980). Etude du processus X n = f(X n − 1) + εn, Comptes Rendus Acad. Sci. Paris, I, 290, 921–923.
Duflo, M. (1997). Random Iterative Models, Springer, Berlin.
Duflo, M., Senoussi, R. and Touati, A. (1990). Sur la loi des grands nombres pour les martingales vectorielles et l'estimateur des moindres carrés d'un modèle de regression, Ann. Inst. Henri Poincaré, Probab. Stat., 26, 549–566.
Gallant, A. R. (1987). Nonlinear Statistical Models, Wiley, New York.
Gallant, A. R. and White, H. (1988). A Unified Theory of Estirnation and Inference for Nonlinear Dynamic Models, B. Blackwell, New York.
Guyon, X. (1995). Random Fields on a Network-Modeling, Statistics, and Applications, Springer, Berlin.
Haggan, V. and Ozaki, T. (1981). Modeling nonlinear random vibrations using an amplititude-dependent autoregressive time series model, Biometrika, 68, 189–196.
Hannan, E. J. (1980). The estimation of the order of an ARMA process, Ann. Statist., 8, 1071–1081.
Haughton, D. (1991). Consistency of a class of information criteria for model selection in nonlinear regression, Comm. Statist. Theory Methods, 20, 1619–1629.
Hertz, J., Krogh, A. and Palmer, R. G. (1991). Introduction to the Theory of Neural Computation, Addison-Wesley, Redwood City.
Klimko, L. and Nelson, P. (1978). On conditional least squares estimation for stochastic processes, Ann. Statist., 6, 629–642.
Lai, T. (1994). Asymptotic properties of nonlinear least squares estimates in stochastic regression models, Ann. Statist., 22, 1917–1930.
Lai, T. and Lee, C. (1997). Information and prediction criteria for model selection in stochastic regression and arma models, Statist. Sinica, 7, 285–309.
Meyn, S. and Tweedie, R. (1993). Markov Chains and Stochastic Stability, Springer, Berlin.
Mokkadem, A. (1987). Sur un modèle autorégressif non linéaire: ergodicité et ergodicité géométrique, J. Time Ser. Anal., 8(2), 195–204.
Nishii, R. (1984). Asymptotic properties of criteria for selection of variables in multiple regression, Ann. Statist., 12, 758–765.
Pötscher, B. M. and Prucha, L. R. (1997). Dynamic Nonlinear Econometric Models, Springer. Berlin.
Quinn, B. G. (1980). Order determination for a multivariate autoregression, J. Roy. Statist. Soc. Ser. B, 42, 182–185.
Schwarz, G. (1978). Estimating the dimension of a model, Ann. Statist., 6, 461–464.
Stout, W. (1970). A martingale analogue of Kolmogorov's law of the iterated logarithm. Z. Wahrscheinlichkeitsthor. Verv. Gebiete, 15, 279–290.
Tjøstheim, D. (1986). Estimation in nonlinear time series models, Stochastic. Process. Appl., 21, 251–273.
Tjøstheim, D. (1990). Nonlinear time series and Markov chains, Adv. in Appl. Prob., 22, 587–611.
Tong, H. (1983). Threshold Models in Nonlinear Time Series, Springer, Berlin.
Tong, H. (1990). Non-linear Time Series (A Dynamic System Approach), Oxford Univ. Press, New York.
Tsay, R. S. (1984). Order selection in nonstationary autoregressive models, Ann. Statist, 12, 1425–1433.
White, H. and Domowitz, I. (1984). Nonlinear regression with dependent observations, Econometrica, 52, 143–161.
Author information
Authors and Affiliations
About this article
Cite this article
Yao, JF. On Least Squares Estimation for Stable Nonlinear AR Processes. Annals of the Institute of Statistical Mathematics 52, 316–331 (2000). https://doi.org/10.1023/A:1004117906532
Issue Date:
DOI: https://doi.org/10.1023/A:1004117906532