Measuring Commodity Correlation with Futures Returns
Summary
The discussion asks whether forward prices should be bootstrapped into estimated spot prices before calculating correlation between two commodities. One response recommends using returns from liquid futures contracts as a practical proxy: calculate daily excess returns for the most liquid maturity, compound them into an index, and then measure correlation at the desired horizon. It also describes checking results with a farther-dated contract. The cited research used this approach in studying time-series momentum and compared near and far contracts, though the discussion gives no commodity-pair correlation estimates.
The main caveat is that futures returns can reflect features of futures markets as well as movements in the underlying commodities. Inferring spot prices from forwards may also be unreliable because spot-forward relationships can behave differently across commodities. The appropriate series therefore depends on the research purpose and the contracts involved; the thread does not establish that bootstrapping is always wrong or offer a complete correlation workflow.
Key ideas
- Futures return series can be used to estimate relationships between commodities without first reconstructing spot prices.
- A common construction uses daily excess returns from the most liquid contract and compounds them into a return index.
- Comparing a farther-dated contract can help assess whether results depend on the selected maturity.
- Futures returns may capture market-specific effects in addition to changes in underlying commodity values.
- Spot-forward behavior can be commodity-specific, so inferred spot prices may not be reliable.
Tags
Full text
# How to calculate correlation between commodities with forward prices? # How to calculate correlation between commodities with forward prices? I'm trying to understand which is the correct way to calculate the correlation between 2 commodities with forward prices. My idea would be first bootstrap forward prices to have only spot prices for each commodity and then to calculate the correlation on these bootstrapped prices. What do you guys think about this idea? would it be too simplistic? Am I forgetting something or this could be a good approach to calculate this correlation? I'm curios to hear your thoughts on this Thanks in advance Nicola ## Answer by cpage (score 2) https://quant.stackexchange.com/a/42808 Drawing on “Time Series Momentum” (Moskowitz, Ooi and Pedersen, 2012), using return series from commodities futures is a perfectly valid way to compute the correlation between the contracts. The danger in this, however, is that you may be picking up effects driven by the peculiarities of the futures markets and not the underlying commodities. To address this, the authors analyze weekly position data from the Commodity Futures Trading Commission (CFTC) to study the trading activity of "speculators and hedgers". From the paper: > We construct a return series for each instrument as follows. Each day, we compute the daily excess return of the most liquid futures contract (typically the nearest or next nearest-to-delivery contract), and then compound the daily returns to a cumulative return index from which we can compute returns at any horizon. For the equity indexes, our return series are almost perfectly correlated with the corresponding returns of the underlying cash indexes in excess of the Treasury bill rate. As a robustness test, we also use the ‘‘far’’ futures contract (the next maturity after the most liquid one). For the commodity futures, time series momentum profits are in fact slightly stronger for the far contract, and, for the financial futures, time series momentum returns hardly change if we use far futures. Additionally, AQR publishes the aggregate commodities return data from the paper here if that's any help. ## Answer by Magic is in the chain (score 1) https://quant.stackexchange.com/a/41484 Depends on the purpose, but I think it would be problematic to infer commodity spots from forward prices. The relationship between the spots and forwards, and the spot price themselves could exhibit very idiosyncratic behaviour, so depends on the commodities as well.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.