Skip to main content Accessibility help
×
Home

Non-stationary q-dependent processes and time-varying moving-average models: invertibility properties and the forecasting problem

  • Marc Hallin (a1)

Abstract

The spectral factorization problem was solved in Hallin (1984) for the class of (non-stationary) m-variate MA(q) stochastic processes, i.e. the class of second-order q-dependent processes. It was shown that such a process generally admits an infinite (mq(mq +1)/2-dimensional) family of possible MA(q) representations. The present paper deals with the invertibility properties and asymptotic behaviour of these MA(q) models, in connection with the problem of producing asymptotically efficient forecasts. Invertible and borderline non-invertible models are characterized (Theorems 3.1 and 3.2). A criterion is provided (Theorem 4.1) by which it can be checked whether a given MA model is a Wold–Cramér decomposition or not; and it is shown (Theorem 4.2) that, under mild conditions, almost every MA model is asymptotically identical with some Wold–Cramér decomposition. The forecasting problem is investigated in detail, and it is established that the relevant invertibility concept, with respect to asymptotic forecasting efficiency, is what we define as Granger-Andersen invertibility rather than the classical invertibility concept (Theorem 5.3). The properties of this new invertibility concept are studied and contrasted with those of its classical counterpart (Theorems 5.2 and 5.4). Numerical examples are also treated (Section 6), illustrating the fact that non-invertible models may provide asymptotically efficient forecasts, whereas invertible models, in some cases, may not. The mathematical tools throughout the paper are linear difference equations (Green's matrices, adjoint operators, dominated solutions, etc.), and a matrix generalization of continued fractions.

Copyright

Corresponding author

Postal address: Institut de Statistique, Université Libre de Bruxelles, C.P. 210, Campus de la Plaine, B 1050 Bruxelles, Belgium.

References

Hide All
Abdrabbo, N. A. and Priestley, M. B. (1967) On the prediction of nonstationary processes. J. R. Statist. Soc. B 29, 570585.
Anderson, O. D. (1978) On the invertibility conditions for moving average processes. Math. Operationsforsch., Ser. Statist. 9, 545–529.
Anderson, O. D. (1977) The time series concept of invertibility. Math. Operationsforsch., Ser. Statist. 8, 399406.
Box, G. E. P. and Jenkins, G. M. (1976) Time Series Analysis. Holden-Day, San Francisco.
Cramér, H. (1961) On some classes of nonstationary stochastic processes. Proc. 4th Berkeley Symp. Math. Statist. Prob. 2, 5778.
Cramér, H. and Leadbetter, M. R. (1967) Stationary and Related Stochastic Processes. Wiley, New York.
Davies, N., Pate, M. B. and Frost, M. G. (1974) Maximum autocorrelations for moving average processes. Biometrika 61, 199200.
De Bruin, M. G. (1974) Generalized C-fractions and a Multidimensional Padé Table. Ph.D. Dissertation, University of Amsterdam.
De Bruin, M. G. (1977) Convergence along steplines in a generalized Padé table. In Padé and Rational Approximation , Academic Press, New York, 1522.
De Bruin, M. G. (1978) Convergence of generalized C-fractions. J. Approximation Theory 24, 177207.
Granger, C. W. J. and Andersen, A. (1978) On the invertibility of time series models. Stoch. Proc. Appl. 8, 8792.
Granger, C. W. J. and Newbold, P. (1977) Forecasting Economic Time Series. Academic Press, New York.
Hallin, M. (1980) Invertibility and generalized invertibility of time series models. J. R. Statist. Soc. B 42, 210212; 43, 103.
Hallin, M. (1981) Nonstationary first-order moving average processes: the model-building problem. In Time Series Analysis , ed. Anderson, O. D. and Perryman, M. R., North-Holland, Amsterdam, 189206.
Hallin, M. (1982a) Nonstationary second-order moving average processes. In Applied Time Series , ed. Anderson, O. D. and Perryman, M. R., North-Holland, Amsterdam, 7583.
Hallin, M. (1982b) Une propriété des opérateurs moyenne mobile. Cahiers du CERO 24 (Mélanges offerts au Professeur P. P. Gillis à l'occasion de son 70e anniversaire) , 229236.
Hallin, M. (1983) Nonstationary second-order moving average processes II: model-building and invertibility. In Time Series Analysis, Theory and Practice 4, ed. Anderson, O. D., North-Holland, Amsterdam, 5564.
Hallin, ?. (1984) Spectral factorization of nonstationary moving average processes. Ann. Statist. 12, 172192.
Hallin, M. and Ingenbleek, J.-Fr. (1981) The model-building problem for non-stationary multivariate autoregressive processes. In Time Series Analysis, Theory and Practice 1, ed. Anderson, O. D., North-Holland, Amsterdam, 599606.
Hallin, M. and Ingenbleek, J.-Fr. (1983) Nonstationary Yule-Walker equations. Statist. Prob. Letters 1, 189195.
Jones, R. H. and Brelsford, W. M. (1967) Time series with periodic structure. Biometrika 54, 403408.
Kendall, M. G. (1971) Book review. J. R. Statist. Soc. A134, 450453.
Levy, H. and Lessman, F. (1961) Finite Difference Equations. Pitman, London.
Magnus, A. (1977) Fractions continues généralisées et matrices infinies. Bull. Soc. Math. Belg. 29B, 145159.
Mattheij, R. M. M. (1980) Characterizations of dominant and dominated solutions of linear recursions. Numer. Math. 23, 421442.
Melard, G. (1962) The likelihood function of a time-dependent ARMA model. In Applied Time Series Analysis , ed. Anderson, O. D. and Perryman, M. R., North-Holland, Amsterdam, 229239.
Melard, G. (1984a) A fast algorithm for the exact likelihood of autoregressive-moving average models. J. R. Statist. Soc. C Appl. Statist. 33, 104114.
Melard, G. (1984b) Analyse des Données chronologiques. Séminaire de Math. Sup., P.U. de Montréal, Montréal.
Miller, K. S. (1968) Linear Difference Equations. Benjamin, New York.
Miller, K. S. (1969) Nonstationary autoregressive processes. IEEE Trans. Inf. Theory 15, 315316.
Milne-Thomson, L. M. (1951) The Calculus of Finite Differences. McMillan, London.
Pagano, M. (1978) On periodic and multiple autoregressions. Ann. Statist. 6, 13101317.
Peiris, S. (1984) Some results on the prediction with nonstationary ARMA models. Statistics Research Report 88, Dept. of Mathematics, Monash University, Australia.
Priestley, M. B. (1965) Evolutionary spectra and nonstationary processes. J. R. Statist. Soc. B 27, 204237.
Rutishauser, H. (1958) Über eine Verallgemeinerung der Kettenbrüche. Z. Angew. Math. Mech. 39, 278279.
Tjöstheim, D. and Tyssedal, J. S. (1982) Autoregressive processes with a time-dependent variance. J. Time Series Anal. 3, 209217.
Troutman, B. M. (1979) Some results in periodic autoregression. Biometrika 66, 219228.
Van Der Cruyssen, P. (1979) Linear difference equations and generalized continued fractions. Computing 22, 269278.
Wall, H. S. (1967) Analytic Theory of Continued Fractions. Chelsea, New York.
Wegman, E. J. (1974) Some results on nonstationary first order autoregression. Technometrics 16, 321322.
Whittle, P. (1965) Recursive relations for predictors of nonstationary processes. J. R. Statist. Soc. B 27, 523532.
Wold, H. (1954) A Study in the Analysis of Stationary Time Series , 2nd edn. Almqvist and Wiksell, Stockholm.

Keywords

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed