Copper cross-market volatility transition based on a coupled hidden Markov model and the complex network method

Abstract

Copper, as a strategic resource, has attracted widespread attention due to its price changes. Copper's futures prices and spot prices tend to be highly correlated in different countries. The copper market price change should consider the relationship between market trends of different countries, especially in the current situation where economic globalization and financial integration are inevitable. The volatility transition between datasets of different countries presents the characteristics of nonlinearity, interaction, dynamics and heterogeneity. To explore the cross-market transition of copper price fluctuations, this paper proposes a hybrid method of the coupled hidden Markov model (CHMM) and complex network. The coupled hidden Markov model (CHMM) is used to identify the transition characteristics of the market price trends, and the complex network method is used to analyze the evolutionary characteristics of the market state transition. We take the copper futures and spot market prices between the U.S., China, and the UK between January 2003 and December 2019 as the object. The results show that the copper futures market's state transition has a clustering effect, the price rise is easy to maintain, and the decline is rapid and easy to transfer. The relationship between daily markets is more volatile, and weekly and monthly are more stable. The British copper futures market has shown advantages in terms of transition ability, assimilation ability, media ability and transfer structure, while the influence of China's copper futures market is relatively weak. In addition, during the financial crisis, the UK spot market played an important role in the frequent declining transition. The nonlinear measurement model and conclusion established in this paper provide a new framework and analysis tool for explaining the volatility transition of the copper market.