The Multi-step Forecasting Model for Price Dynamics of Global Commodity Markets

This work is devoted to methods of short-term forecasting of prices on the world commodity markets. The main goal of this work is to build a methodology for short-term forecasting of spot prices for ammonia in twelve different regions of the world. To achieve this goal, the following tasks are set. First, we need to analyze the main approaches to short-term time series forecasting. Second, identify a list of potential exogenous regressors and collect relevant statistical information on them. Third, determine the range of predictive models and possible methods from the ensemble. Fourth, implement the generated approach in the R programming language. Fifth, make the necessary calculations and measure the quality of forecasts. Conclusions based on the results obtained will contribute to the construction of a complete methodology for predicting prices for ammonia.