ASSESSMENT OF THE QUALITY OF THE STATISTICAL FORECAST FOR THE SHARE PRICE

The stock market is an important part of the world's economy, but values stocks change at a significant speed and finding effective method forecasting price will be victory for investors. article devoted to use statistical methods in market, evaluating their effectiveness on example determining forecast share comparing it with real price. In addition, work was carried out determine relevance research topic by means bibliographic analysis based results searching selection publications Scopus database subsequent VOSviewer software toolkit. A time series covering period daily prices Apple Inc. chosen as input test sample developing forecast. from December 12, 2016 2021. At same time, median monthly financial shares com corporation were used build models, since there no information that fell weekends studied period. Forecasts are published using Statgraphics 19 software, which generates 17 models: random walk, walk drift, constant mean, linear trend, quadratic exponential S-shaped simple moving average, smoothing coefficient α = 0.9999, Brownian factor α= 0.5705, Holt coefficients 0.9999 β 0.0359, 0.3046, ARIMA model. Akaike's criterion assess quality models. As result analysis, ARIMA(1,0,0) model has lowest value according criteria. On basis ARIMA(1,0,0), exchange rates built 6 months ahead age final array. process comparison data, discrepancy between forecasted found, however, obtained models tested significance. conducted showed possible classical develop only short-term period, while necessary take into account risk factors inherent (the frequency changes during day, liquidity risks, operational political situation world).