Commodity skewness
PortfolioAdvanced
A commodity skewness sort ranks futures on the asymmetry of their recent daily returns. The researched rule buys the contracts with the most negative skew and shorts the contracts with the most positive skew. The economic story is that investors overpay for lottery-like upside in commodities. It is a cross-sectional risk sort, separate from momentum and from the shape of the futures curve.
Not investment advice. Nothing here is a recommendation to buy or sell anything. Backtests and historical studies are not guarantees, and a rule that worked in one sample can fail in the next. Trading can lose money. Links are for education, not endorsements. Do your own research.
- Rules
- Each month, compute a stated skewness statistic on a trailing window of daily returns for each commodity future. Buy the lowest-skew group and short the highest-skew group in equal weight. Rebalance monthly. Use excess returns after the roll. Keep the window and the group size fixed. Do not drop a commodity because its skew looks odd unless a liquidity rule says to.
- Where AI fits
- AI can compute skewness on the rolled series, handle missing days the same way every month, and show the correlation of this book with curve-carry and momentum books. High correlation means the 'new' sort is an old bet. The window length has to be chosen out of sample.
- Markets
- A diversified set of liquid commodity futures.
- Timeframes
- Monthly rebalance. Skewness is estimated on daily data over roughly a year in typical designs.
- Risks
- Skewness estimates are unstable and dominated by one shock day. The short lottery leg can pay out all at once. Small commodities cannot absorb the short. Overlap with other commodity premia means the risks are not additive diversification. A historical sort is not an expected-return quote.