This paper tests a wide range of momentum and reversal strategies at different trading frequencies for the complete Chinese commodity futures market dataset. Accurate estimates of transaction costs for each commodity and the minute level futures prices are utilized to obtain the most realistic out-of-sample backtesting results. Distinctively from the existing literature, our dataset does not suffer from liquidity problems since the intra-day data is constructed from the most actively traded contracts for each and every of the 31 commodities included in our sample. Overall, there are three main findings of this study. First, momentum and reversal trading strategies can generate robust and consistent returns over time; however, the intra-day strategies used cannot generate sufficiently enough high excess returns to cover the excessive costs due to the higher frequency of trading. Secondly, at lower trading frequencies and longer holding periods momentum and reversal strategies can generate excess returns, but with higher maximum drawdown risk. Finally, the double-sort strategies statistically improve the performance of the trading strategies.
- Chinese commodity futures market
- Inter-day and intra-day frequencies
- Single-sort and double-sort strategies