Financial contagion, price discovery process and hedging effectiveness in the futures markets : evidence from wavelet analyses

Abstract

This thesis examines the dynamics of futures markets from the time-frequency perspective. Our dataset consists of forty futures contracts and underlying spot prices worldwide, spanning from 2010 through to 2020. The objectives of the thesis are three-fold. First, we examine the occurrence of financial contagion and its duration in the futures markets using the Wavelet Correlation Breakdown test and Wavelet Power and Energy Spectrum scalogram. The results confirm the existence of financial contagion in the futures markets and that the contagion lasts for about two months. Second, we examine the multiscale spot-futures co-movement and their causality directions by applying the Wavelet Correlation, Wavelet Coherence and Wavelet Phase Difference. We also conduct a couple of robustness tests using the Dynamic Conditional Correlation (DCC-GARCH) and Toda-Yamamoto approach of the Granger Causality test. The results of the main and robustness methods show that the agricultural and energy futures markets are less efficient for the short-term price discovery process. The weak short-term positive correlation causes temporary spot-futures price divergence, thus providing a diversification strategy for investors who seek short-term profit opportunities from the futures markets. Finally, for the third objective, we investigate the multiscale optimal hedge ratio and hedging effectiveness of the futures markets by using the Wavelet Variance-Covariance and Wavelet Squared Correlation analyses. We also employ the Vector Error Correction (VEC) model and Minimum Variance Hedge Ratio (MVHR) procedure as the robustness checks. Our findings from both the main methods and the robustness checks indicate that stock index and metal futures contracts have excellent hedging effectiveness in both short- and long-term hedging horizons. Generally, our thesis suggests three main trading strategies for both hedgers and non-hedgers. First, the hedgers should not hold hedging positions for less than two months due to weak positive short-term correlation. Second, due to the weak positive short-term spot-futures correlation, the non-hedgers may gain from short-term profit opportunities throughout the multiple trading activities in the agricultural and energy futures markets, should they decide to speculate. Finally, the stock index and metal futures contracts appear to be the best performers in the price discovery process. They serve as very effective hedging pairs to their underlying assets since their spot-futures correlations are highly positive at all scales. By and large, the thesis adds to the growing financial derivatives literature by helping investors make informed investment decisions, particularly in formulating hedging and asset allocation strategies.