Ситуаційно-індуктивне моделювання в задачах екстраполяційного прогнозування за даними моніторингу

• Datum: 2016
• 1. Verfasser: Stefanyshyn, Dmytro V.
• Format: Artikel
• Sprache: Ukrainisch
• Veröffentlicht: The National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute" 2016
• Schlagworte: екстраполяція; індуктивні та ситуаційні моделі; моделювання та прогнозування за даними моніторингу; оперативні і строкові прогнози
• Online Zugang: http://journal.iasa.kpi.ua/article/view/70651
• Назва журналу: System research and information technologies

The approach to forecasting phenomena and processes with extrapolations using monitoring data and based on the idea of situational-inductive modelling is proposed. According to this approach, results of situational modelling within the samples time series, which are characterized by the monotony of trends or quasi stationary behaviour of variables of models at the appropriate time intervals, form the basis for the subsequent construction of inductive models. Inductive models are performed as models of "levels" which are based on the results of the situational modelling. Herewith, on the basis of situational models within intervals where situational models are considered adequate, the operational forecasting (in real time) is carried out, and on the basis of inductive models by means of which within the observed data the evolution of situational models is controlled, the term forecasting is carried out to establish situational models for future time periods.