Advertisement

Geography and Natural Resources

, Volume 39, Issue 1, pp 73–78 | Cite as

Quality and Reliability Parameters of Predictive Models in Hydrometeorology

  • А. V. Ignatov
Research Techniques
  • 9 Downloads

Abstract

This paper discusses the main properties of models and model assessments forming the notion of their quality, which include the accuracy, reliability and details of the description of approximate model assessments. On the basis of taking into account the entire set of quantitative parameters of these characteristics, the notion of the measure of model quality is introduced. The main property determining the model quality is highlighted, namely the accuracy of model calculation using independent data. Approximate estimations of the values of a variable in the interval and probabilistic form are considered, which can be constructed on the basis of data on its values in the past. For probabilistic assessments of the values of the variables, it is suggested that, in addition to the measure of accuracy, the measure of informativity should be used, which is determined in terms of entropies of the corresponding probability distribution functions. In developing the algorithms for assessing the measure of quality, special attention is paid to calculating the parameters characterizing the reliability of approximate model assessments and the models themselves. The reliability is assessed in terms of the probabilities of the events occurring when the model is constructed or used. The measure of probability of the model of dependence determined at the stage of constructing it by using the learning sample is calculated as the product of two probabilities: the measure of confidence to the predictors of a dependent variable, and the measure of confidence to the operator describing a dependence of this variable on them. Recommendations are made for assessing these probabilities. In view of the stricter requirements for the reliability of predictive assessments, the algorithm is suggested for increasing it by combining into an ensemble the conditional model and unconditional probability assessments.

Keywords

construction of models verification of hypotheses properties of approximate assessments measure of model quality improvement in predictability 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kattsov, V.M., Climate Prediction: Progress, Problems, and Prospects, Russ. Meteorol. Hydrol., vol. 35, issue 1, pp. 10–12.Google Scholar
  2. 2.
    Kendall, M.G., and Stuart, A., The Advanced Theory of Statistics, in 3 Vols., Vol. 3, Design and Analysis, and Time-Series, London: Charles Griffin Company Limited, 1966.Google Scholar
  3. 3.
    Kondratiev, K.Ya., Uncertainties of Observation Data and Numerical Climate Simulation, Meteorol. Gidrol., 2004, no. 4, pp. 93–119 [in Russian].Google Scholar
  4. 4.
    Kuzmin, V.A., Selection and Parametrization of Prediction Models of River Flow, Meteorol. Gidrol., 2001, no. 3, pp. 85–90 [in Russian].Google Scholar
  5. 5.
    Khristoforov, A.V., Concerning the Problem of Forecasting Hydrological Characteristics From Regression Equations, Meteorol. Gidrol., 1975, no. 11, pp. 72–80 [in Russian].Google Scholar
  6. 6.
    Khristoforov, A.V., Stability of Occurrence of Linear Correlations in Hydrological Calculations and Forecasts, Vestn. Mosk. Univ., Ser. Geogr., 1982, no. 5, pp. 56–61 [in Russian].Google Scholar
  7. 7.
    Velikanov, M.A., Errors of Measurement and Empirical Dependencies, Leningrad: Gidrometeoizdat, 1962 [in Russian].Google Scholar
  8. 8.
    Kazakevich, D.I., The Foundations of the Theory of Random Functions and Its Use in Hydrometeorology, Leningrad: Gidrometeoizdat, 1977 [in Russian].Google Scholar
  9. 9.
    Alefeld, G. and Herzberger, J., Introduction to Interval Analysis, New York: Academic Press, Inc., 1983.Google Scholar
  10. 10.
    Ignatov, A.V., The Idea of the Algorithm of Extrapolating the Nonstationary Hydrometeorological Series Using an Optimized Empirical Model, Geogr. Nat. Resour., 2010, vol. 31, issue 1, pp. 68–73.CrossRefGoogle Scholar
  11. 11.
    Ventsel’, E.S., Probability Theory, Moscow: Gos. Izd. Fiz.-Mat. Literatury, 1962 [in Russian].Google Scholar
  12. 12.
    Ignatov, A.V., Information-Probability Simulation in Geographic Research, Geogr. Prir. Resur., 2006, no. 2, pp. 27–33 [in Russian].Google Scholar
  13. 13.
    Shannon, C., Works on Information Theory and Cybernetics, Moscow: Izd. Inostr. Literatury, 1963 [in Russian].Google Scholar
  14. 14.
    Korn, G.A. and Korn, T.M., Mathematical Handbook for Scientists and Engineers, 2nd Ed., New York: McGraw-Hill, 1968.Google Scholar
  15. 15.
    Reznikov, A.P., Prediction of Natural Processes by the Learning System, Novosibirsk: Nauka, 1982 [in Russian].Google Scholar
  16. 16.
    Ignatov, A.V., Kravchenko, V.V. and Chekmarev, A.A., Stochastic Modeling: “Expert” Version. Certificate of State Registration of Computer Software No. 2016617121 of 6.27.2016. URL: http://www.irigs.irk.ru (Accessed 3.31.2017) [in Russian].Google Scholar
  17. 17.
    Burakov, D.A., Gordeev, I.I., Ignatov, A.V., Petkun, O.E., Putintsev, L.A., and Chekmarev, A.A., Forecasting the Water Inflow Into the Krasnoyarsk and Sayano-Shushenskoe Reservoirs in the Second Quarter of the Year, Geogr. Nat. Resour., 2016, vol. 37, issue 2, pp. 158–164.Google Scholar

Copyright information

© Pleiades Publishing, Ltd. 2018

Authors and Affiliations

  1. 1.V. B. Sochava Institute of Geography, Siberian BranchRussian Academy of SciencesIrkutskRussia

Personalised recommendations