Futures Price Series: Methodology Matters

Oil - Oil jackpump at sunset 2 by Evgenii Mitroshin via iStock

Brett Friedman, Winhall Risk Analytics/OptionMetrics Contributor

 

If you are trading futures, you should know something important. Continuous futures price series may vary between sources and cause misleading analysis. The culprit here is how the various contract months, each with a different price and expiration, are combined to construct a long term price series or chart. Since they can be used for back testing and to determine trading strategies, portfolio risk, metrics, and trading limits, it is extremely important to understand how futures series are constructed and the inherent compromises that could occur. 

 

The root of the problem, multiple contract months with different prices and expiration dates, is easily understood. Since contract months are either positively sloped (contango) or negatively sloped (backwardated), a continuous price series will show an increase or decrease when the front month contract expires and the next month takes over. For example, on 08/20/24, the front month WTI crude oil contract (September) expired and settled at $74.04. On 8/21/2024, the new front month contract, October, settled at $71.93. The question then becomes how to construct a continuous series, and more specifically, how to deal with the roll effect without introducing undue distortion into the continuous series.  

 

The most common methodology, front monthly only, is outlined above. Simply, the continuous series is constructed from whichever contract represents the front month on any date. Most continuous price charts available online and in the popular financial press utilize this technique. It is simple and retains the integrity of the original prices without smoothing or manipulation. However, in situations in which the futures curve is steeply sloped, and significant differences exist between the individual contract months, the results might be misleading. In this case, the daily change may produce extreme daily volatility, or even a daily gap, when it is really more a result of extreme contango or backwardation.  

 

Numerous methodologies attempt to account for this problem, with varying degrees of success. By far the most common, and well-used in banks and hedge funds, constant maturity future (CMF) methodology involves interpolating prices from several different contract months to produce a constant, fixed maturity. For example, a 30-day CMF would involve an interpolation of the first and second contract months. Each day in the series would then have 30 days to expiration. Longer constant maturities would involve interpolating additional contracts (assuming that longer maturity contacts are available and liquid). In this manner, expiration-related volatility is eliminated, and cyclicality and seasonality are reduced. 

 

Of course, the constant maturity future methodology presents new problems. The unique features of each contract are blended away, and the increasing risk as nearby contracts approach expiration is understated. In addition, if historical worst-case limit metrics are employed (i.e., what is the worst case historically for a single position or portfolio?), then this method obscures the true historical results by using a theoretical and ahistorical price series that ignores seasonality. The latter is especially important in commodities that display significant seasonality, such as natural gas and grains. In addition, since the interpolated prices change each day in order to maintain the constant maturity, it is difficult to predict which period represents the worst result, and it can shift significantly from one day to the next, producing counterintuitive results. More broadly, for limits based on historical events or scenarios, using a methodology that produces manipulated and ahistorical data is contradictory. Applying a single maturity futures series to a specific position (i.e., “what is the worst that ever happened to the August contract month, historically?”) would then produce clearer and more consistent results, albeit with less data.

 

Front month only and continuous futures methodologies are compared below:

Sept 2024 Newsletter Copy V1 2docx Docx
Mail Liah Jimenez Outlook

As you can see, although the general shape of the curves is very similar, important details vary. For example, for April 20, 2020, perhaps the most notable day in the history of crude oil trading, the front month only chart shows a settlement of -$37.63, while the constant maturity futures (30-Day) shows $20.60. Not only is the negative settlement missing from the CMF chart, but the two settlements show a difference of $58.23!  

 

Ultimately the proper continuous futures methodology depends on the desired application. If the object is to analyze front month to second month roll patterns over time, or to chart historical price patterns and trends, then a continuous front month only methodology is most appropriate and simple to construct. However, if the series is being used to determine a position’s risk metrics, VaR, or limits, then the front month only methodology may introduce misleading roll gap-related volatility, thereby possibly overstating risk. Similarly, constant maturity future methodology, especially when used to determine daily limits, can produce confusing results.  

 

Unfortunately for traders and risk managers, there is no one-size-fits-all solution. Besides understanding exactly which methodology is being employed for existing price series, a flexible database is essential to construct futures time series appropriate to the desired application.  

 

For more insights on the commodities and futures, the options market, and volatility, visit the OptionMetrics blog.

There is an inherent risk involved with financial decisions. The information in this article is for informational purposes only and is not intended to provide financial advice.

 

 


On the date of publication, OptionMetrics did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. For more information please view the Barchart Disclosure Policy here.