OPTIKA ATMOSFERY I OKEANA JOURNAL
V.E. ZUEV INSTITUTE OF ATMOSPHERIC OPTICS
Vol. 3, issue 01, article # 10
pdf Kalinenko A. N., Teushchekova T. G. Methods of self-organization of models in problems of physical-statistical prediction of altitude profiles of the temperature. // Atmospheric and oceanic optics. 1990. V. 3. No. 01. P. 50-56.Copy the reference to clipboard
Abstract:
Possible applications of algorithms of the method of grouping of arguments (MGA) in problems of medium- and long-term prediction of the altitude profiles of the temperature are discussed. These algorithms make it possible to construct simple predictive models with optimal complexity based on a special table of the initial experimental data (predictors), which hold a long history of the process, and with the help of criteria set by experts. The effectiveness of this approach in medium-term prediction of the altitude profiles of the temperature is demonstrated for a concrete example.