On the macroeconomic fundamentals of long-term volatilities and dynamic correlations in COMEX copper futures

TL;DR

This study investigates how macroeconomic variables influence copper futures volatility and their long-term correlation with the S&P 500, using advanced GARCH-MIDAS and DCC-MIDAS models to identify key drivers and improve modeling accuracy.

Contribution

It introduces the application of GARCH-MIDAS and DCC-MIDAS frameworks to analyze macroeconomic impacts on copper futures and demonstrates the effectiveness of PPI in long-term correlation modeling.

Findings

PPI is the most influential macroeconomic variable on copper futures returns.

MIDAS filter enhances model performance over traditional RV.

Long-term correlation with S&P 500 is better captured using PPI with MIDAS filtering.

Abstract

This paper examines the influence of low-frequency macroeconomic variables on the high-frequency returns of copper futures and the long-term correlation with the S&P 500 index, employing GARCH-MIDAS and DCC-MIDAS modeling frameworks. The estimated results of GARCH-MIDAS show that realized volatility (RV), level of interest rates (IR), industrial production (IP) and producer price index (PPI), volatility of Slope, PPI, consumer sentiment index (CSI), and dollar index (DI) have significant impacts on Copper futures returns, among which PPI is the most efficient macroeconomic variable. From comparison among DCC-GARCH and DCC-MIDAS model, the added MIDAS filter of PPI improves the model fitness and have better performance than RV in effecting the long-run relationship between Copper futures and S&P 500.