EFORECASTING FINANCIAL INDEXES WITH MODEL OF COMPOSITE EVENTS INFLUENCE

In this article we propose the model for the forecast of various financial indexes: stock markets indexes; currency exchange rates; crediting rates. Behaviour of financial indexes depends on psychological sentiments of players (investors, traders) and their inclination to buy or sell financial tools. We have made the supposition that political, economical, financial and other events are preconditions for formation of the future psychological sentiments of players. Therefore, for forecasting financial indexes we estimate influence of all topical events on the future inclination of players to buy or sell. The proposed model calculates the composite influence of events on the basis of estimations of influence direction, influence force, influence time, events importance and confidence to the information about events. The model fulfils the calculations with help of fuzzy integral Sugeno (1972). We have used this model for forecasting indexes of various economical natures: Ukrainian stock index (PFTS); exchange rate EUR/USD; crediting rate KievPrime 1M and quotations of Eurobonds Ukraine 2015. We also have estimated errors and horizons of forecasts..