Building Constant-Maturity Futures Series from Listed Contracts
Summary
The document asks how to build a five-month continuous gold futures series from individual contracts with different listing and expiry dates, in the context of replicating commodity models. It distinguishes the mechanics of stitching nearby contracts from the separate choice of weighting method, but does not give a roll schedule or explain how to interpolate between contracts.
The response points to Refinitiv's composite commodity futures continuation series, identifying maturity points at one, five, nine, thirteen, and seventeen months as the source used to form the table in question. This offers a practical data-series solution for replication, rather than a general construction recipe. The answer does not specify the vendor's roll rules, adjustments, or treatment of missing observations, so users seeking to build equivalent series from raw contracts would need further documentation.
Key ideas
- Continuous futures data at specified maturities can support replication of commodity pricing models.
- The example uses Refinitiv composite continuation series at regular four-month maturity intervals.
- The response identifies a data product but does not explain its roll or stitching methodology.
Tags
Full text
# How to construct continuous futures contracts with multiple maturities # How to construct continuous futures contracts with multiple maturities I am trying to replicate the Schwartz-Smith (2000) model and having an issue understanding what the data is and how to generate it. Specifically, the authors use a table of continuous futures with expirations maturity = [1/12,5/12,9/12,13/12,17/12]. How exactly is this created? I have daily prices for a series of gold futures: GCM1, GCZ7, GCZ9, GCM9, GCV8, GCQ7 for example, but these contracts all have a start and end date. What is the way to create the 5-month continuous futures contract from this data? This was asked in Where can one find the daily prices of commodity futures of multiple maturities and time to expiration of the contracts? but the answers were about the data sources, not the mechanics of stitching the contracts together. Additionally, I can find resources about the different weighting methods when stitching two contracts together, but nothing about how to assemble the time series. ## Answer by user86422 (score 1) https://quant.stackexchange.com/a/78549 I'm still unsure about a general approach to this, or how to answer my specific question about how to create a five month continuous contract from a series of futures. But, in order to replicate the data found in the Schwartz-Smith (2000) or Schwartz (1997) papers, I needed to find a Composite Commodity Future Continuation. In this case, I used the Refinitiv GCc1, GCc5, GCc9, GCc13, and GCc17 data series to create my table of continuous futures with expirations maturity = [1/12,5/12,9/12,13/12,17/12].
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.