Blog - Investing Notes

July 31, 2007 - Does Technical Trading Work with Commodity Futures?

Do relatively low transaction costs and ease of short selling enable profitable technical trading in commodity futures markets? In their recent paper entitled "Can Commodity Futures be Profitably Traded with Quantitative Market Timing Strategies?", Ben Marshall, Rochester Cahan and Jared Cahan investigate the effectiveness of 7,846 quantitative trading rules from five rule families (Filter, Moving Average, Support and Resistance, Channel Breakouts, and On-Balance Volume) for 15 kinds of commodity futures contracts. They test these rules for cocoa, coffee, cotton, crude oil, feeder cattle, gold, heating oil, live cattle, oats, platinum, silver, soy beans, soy oil, sugar and wheat futures. Their testing includes two bootstrapping methodologies, adjustment for data snooping bias and evaluations over different time periods. Using daily price and volume data for 1984-2005, they conclude that:

• Across the entire sample, nine of 15 commodities have positive mean daily returns, and all exhibit positive kurtosis (fat tails).
• The best trading rule for each commodity yields statistically significant profits. However, adjustment for data
  snooping bias eliminates reliable profits for 14 of 15 commodities.
• The best moving average rule generates profits in excess of reasonable transactions costs for oats even after adjusting for data snooping bias. However, this profitability disappears in the 1995–2005 subperiod.
• Overall, results indicate that technical rules are not profitable as standalone commodity futures trading methods, but they do not rule out the possibility that such rules might compliment some other trading strategy.

In summary, market timing based on technical analysis is not reliably profitable for commodity futures.

For related research, see Blog Synthesis: Investing/Trading in Commodities and Commodity Futures and Blog Synthesis: Some Trading Indicators. See especially our blog entry of 4/25/07 on a very similar study of Standard and Poor’s Depository Receipts using high-frequency intra-day data by the same authors.