Physical Oceanography

, Volume 17, Issue 4, pp 242–251 | Cite as

Prediction of time series by the method of analogs

  • E. F. Vasechkina
  • V. D. Yarin


We consider an algorithm of prediction of nonstationary time series based on the method of analogs. Since the exhaustion of a great number of versions is required for the adjustment of the parameters of the optimal prognostic model, we describe a genetic algorithm used in this case. We consider several procedures of construction of prognostic models. The numerical results are used to choose the procedure guaranteeing the minimum mean square error. The parameters of the model affecting the quality of predictions are determined. The proposed method is tested by using the reanalysis data (NCEP/NCAR project) on the anomalies of the monthly average surface air temperature for 58 yr. The results of predictions are compared with the estimates obtained by the linear regression method. It is shown that the method of analogs gives satisfactory results even in the cases where the regression methods lead to errors equal to the variance of predicted series.


Genetic Algorithm Neighboring Node Temperature Anomaly Empirical Orthogonal Function Reanalysis Data 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    E. F. Vasechkina and V. D. Yarin, “Genetic algorithm in the problem of reconstruction of hydrometeorological fields,” in: Systems of Monitoring of the Environment-2001 [in Russian], É KOSI-Gidrofizika, Sevastopol (2002), pp. 141–145.Google Scholar
  2. 2.
    E. F. Vasechkina and V. D. Yarin, “Application of the genetic algorithm to the problem of reconstruction of missing data,” Morsk. Girdogiz. Zh., No. 4, 30–39 (2002).Google Scholar
  3. 3.
    H. M. Van Den Dool, “A new look at weather forecasting through analogues,” Month. Wea. Rev., 117, October, 2230–2247 (1989).Google Scholar
  4. 4.
    NCEP Reanalysis Data. NOAA-CIRES ESRL/PSD Climate Diagnostics branch, Boulder, Colorado, USA;
  5. 5.
    D. Beasley, D. R. Bull, and R.R. Martin, “An overview of genetic algorithms: Part 1: Fundamentals,” University Comput., 15(2), 58–69 (1993).Google Scholar

Copyright information

© Springer Science+Business Media, Inc. 2007

Authors and Affiliations

  • E. F. Vasechkina
    • 1
  • V. D. Yarin
    • 1
  1. 1.Marine Hydrophysical InstituteUkrainian Academy of SciencesSevastopol

Personalised recommendations