본문 바로가기
  • Home

A Study on Early Warning System of China Iron Ore Price Based on AI Models

Nam, Dae-yub 1

1포스코경영연구소

Accredited

ABSTRACT

AI(Artificial Intelligence) has been attracting much attention since the coming out of Alpha Go in 2016. Recently, Economic studies have been actively conducted using AI-based machine-learning and deep-learning models instead of classical time series or regression models. In this paper, I applied the AI technique to the EWS(Early Warning System) for Chinese iron ore import price, which has an important influence on the sound development of Chinese steel industry. The work presented in this paper aims to examine the comparison analysis of the prediction accuracy between the classical signal approach which are mainly used in the contemporary economics and the AI-based SVM(Support Vector Machine) and Random Forest models. And I described the pros and cons of each technique. The results show that the AI models have achieved higher accuracy than signal approaches. So, it is reasonable to think that AI models are more suitable than the traditional models for studies which focus on model’s accuracy, such as EWS. And these models can help steel companies prevent volatility in financial performance by increasing iron ore inventory or using of financial derivatives.

Citation status

* References for papers published after 2022 are currently being built.