Karen G. Paytyan

Candidate for a Degree, Department of Applied Informatics and Mathematical Methods in Economics, Volgograd State University, Prosp. Universitetsky, 100, 400062 Volgograd, Russian Federation; Analyst, Analytical Reports Division, Analytics and Risk Management Department, EOS Group, EOS LLC, Tverskaya St, 12, Bld. 9, 125009 Moscow, Russian Federation, This email address is being protected from spambots. You need JavaScript enabled to view it. ,

Abstract.  The volume of world metal consumption is one of the main indicators of the state of economy as a whole. This is explained by the fact that such an industry as construction presents a great demand for these products. At the same time, the volume of construction is growing along with the economy growth, because a healthy market attracts more investment. Therefore, the state of the global metal market is one of the indicators of the state of the global economy as a whole, and the challenges facing this industry are relevant for the entire world market. One of them is forecasting metal prices to make right business decisions. The article presents a practical task that shows the need for forecasting. At the next step, the author developed a criterion for the quality of forecasting, if satisfied, we can talk about the applicability of the model in practice. On a randomly selected time interval, the quality of common statistical forecasting models, such as the pair regression equation and linear models, is analyzed. New models have also been developed that are based both on a technical analysis of exchange quotations of metal prices and on a fundamental one. At the final step, the results of all the models presented in the work were compared with the criterion of applicability developed in practice by the author and the most promising of them were selected.

Key words: standard statistical models for forecasting time series, “slow” and “fast” moving averages, futures contract price, fair futures price, actual futures price, forecast model quality criterion, proportion of correctly predicted price directions, average relative forecast error.

Creative Commons License

DEVELOPMENT OF FORECASTING MODELS SUITABLE FOR METALTRADING COMPANIES by Paytyan K.G. is licensed under a Creative Commons Attribution 4.0 International License.

Download this file (3_Paytyan.pdf) 3_Paytyan.pdf
538 Kb419 Downloads