Summary
First, the Group Method of Data Handling (GMDH) algorithm, a self-organizing method for constructing higher order regression, is applied to build a model for long-term forecasting. Then, the Group Method of Phase Space Component (GMPSC) model is set up based on chaos theory and GMDH. Several case studies show that both GMDH and GMPSC provide an efficient and potentially useful tool for non-linear time series modeling.
Similar content being viewed by others
References
Grassberger, P., Procaccia, I., 1983: Measuring the strangeness of strange attractors,Physica,9D, 189–208.
Henderson, H. W., Wells, R., 1988: Obtaining attractor dimension from meteorological time series.Adv. Geophys.,30, 205–237.
Havstad, J. W., Ehlers, C., 1989: Attractor dimension of non-stationary dynamical systems from small data sets.Phy. Rev.,A39, 845–853.
Ivakhnenko, A. G., Ivakhnenko, N. A., 1974: Long-term prediction by GMDH algorithm using the unblanced criterion Sov.Autom. Control.,7(4), 40.
Lin, Z.-S., 1991: The phase space theory and models of long-term forecast. Ph.D. Dissertation, Peking University.
Ruelle, D., 1990: Deterministic chaos: the science and the fiction.Proc. R. Soc. Lond.,A427, 241–248.
Author information
Authors and Affiliations
Additional information
Supported by the NKPFR “Climate Dynamics and Climate Prediction Theory”.
Rights and permissions
About this article
Cite this article
Lin, Z.S., Liu, J. & He, X.D. The self-organizing methods of long-term forecasting (I) — GMDH and GMPSC model. Meteorl. Atmos. Phys. 53, 155–160 (1994). https://doi.org/10.1007/BF01029610
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01029610