The paper is aimed at commodity price forecasting using a numerical solution of the Cauchy initial problem for the 1st order ordinary differential equation. To acquire significant forecasting improvement, the idea of the modification of the initial condition value was realized. By having analysed the forecasting success of determined numerical models, it was found out that commodity price evolution affected the accuracy of the price forecasting. The absolute percentage prognoses errors were usually lower at a stable price increase and when price fluctuation appeared. Therefore, prognoses calculated without changing initial condition value were satisfying. Within significant changes in the price evolution and at a rapid price increase, the prognoses acquired higher absolute percentage errors. That caused replacing the initial condition value by the nearest stock exchange. Consider this strategy, the following calculated prognoses got closer to the forecast stock exchanges and price forecasting became more advantageous with respect to the price course.


    Access

    Download


    Export, share and cite



    Title :

    INFLUENCE OF THE PRICE MOVEMENTS TO THE ACCURACY WITHIN NUMERICAL PRICE FORECASTING


    Contributors:


    Publication date :

    2019




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    Unknown