Empirical Analysis of Copper Co-movement, Volatility and Hedge Ratios with Top Producing Countries

Abstract

Listen

Translate article

An empirical analysis of copper and a selection of currencies and indices of the countries that produce the largest amount of copper is done in the study to evaluate co-movement, volatility effects and static hedge ratios. The countries that are included are Chile, Peru, China, Australia, Congo, Zambia, Mexico, Indonesia, Canada, Russia, Poland, Kazakhstan, Brazil, South Africa and India. An index for Congo was not available, but the currency was included. To evaluate the overall effect on copper to emerging markets, the MSCI Emerging Market 50 Index was also included as part of the variables. The data will be evaluated by means of number of financial econometric models. The methodology includes correlation, VAR, Johansen Cointegration, Granger Causality, the generalised autoregressive conditional heteroscedastic (GARCH) model, Glosten-Jagannathan-Runkle generalised autoregressive conditional heteroscedastic (GJR-GARCH) model, and the exponential GARCH (EGARCH) model. The final part of the analysis includes the graphical representation of dynamic conditional correlations and four static hedge ratios which are the OLS methodology, ECM methodology, VECM methodology and finally the ECM-GARCH methodology. The models will be based on daily data from 1 August 2011 to 9 April 2018. The Granger Causality results suggest that relationships exist between all the variables except for seven currencies and one index, namely, the currencies for Australia, Brazil, Canada, Kazakhstan, Peru, Poland and Congo. The Index that does not show any relationship is the Zambian Lusaka All share Index. Volatility is present in the data and therefore the models mentioned will be compared in order to identify which model is the best fitting model for the selected commodities, currencies and index. Overall, GJR-GARCH was the best fitting model for copper spot and future, in addition, leverage effects exist which imply that negative shocks have a greater effect than positive shocks. The static hedge ratio analysis showed that Russian RPS Index and copper future provided the largest value, followed by the Brazilian Bovespa Index and the Peruvian S&P/BVL General Index. On the negative values, the largest negative value was obtained for the South African FTSE/JSE All Share Index.