Forecasting the inflation rate based on big data in the context of implementing import substitution policies
Abstract. This article considers inflation forecasting methods using big data analysis in the context of import substitution policy implementation. The relevance of the study is determined by the need to adapt economic policy to modern challenges associated with global changes and internal economic processes. The authors scientifically substantiate that the use of big data analysis methods can increase the accuracy of forecasts and improve management decisions, which in turn contributes to sustainable economic development. Scientific novelty lies in the adaptation of forecasting methods to the specifics of an economy focused on import substitution, this includes taking into account changes in the structure of production, supply chains and consumer preferences, which has not previously been considered within the framework of classical inflation forecasting models. The results of the study can form a basis for developing more accurate and adaptive tools for managing inflation processes in the context of economic transformation.
Keywords: inflation, big data, import substitution policy, FAVAR models, DMS and DMA models, machine learning methods, random forest algorithm, gradient boosting algorithm.
Hightlightts:
- the theoretical justification for the need to study inflation forecasting based on big data in the context of import substitution policy implementation is presented; the feasibility of the following assessment methods is substantiated: FAVAR models, DMS and DMA models, machine learning methods (random forest algorithm, gradient boosting algorithm);
- inflation forecasting using various assessment methods with subsequent determination of forecast quality is presented;
- a comparative analysis of the models among themselves in terms of quality relative to the reference linear regression model is carried out.
Yuri E. Alexandrovich, Irina A. Eremina - Peter the Great St. Petersburg Polytechnic University, St. Petersburg, Russia