RollFutures

Code to download futures contract data (primarily via Quandl) and construct a continuous nearby futures series. Supports multiple rolling methodologies.

Features

• Constructs a continuous nearby futures series from individual contracts.
• Supports three primary rolling logic methods:
  • Volume-based (roll_method = "volume"): Rolls to the next contract in the series when its daily volume exceeds the volume of the currently active contract.
  • Days to Expiration (roll_method = "dte"): Rolls to the next contract a specified number of days before the approximate expiration of the current contract.
  • Open Interest-based (roll_method = "oi"): Selects the contract with the highest daily open interest as the nearby contract.
• Downloads historical data using the Quandl API (specifically targeting CME data).
• Generates a plot of the selected nearby series using ggplot2.
• Exports the selected nearby series to a CSV file with a dynamic filename.

Configuration

Open RollFutures.R and modify the parameters in the "User-Configurable Parameters" section at the top of the script:

• Quandl.api_key("YOUR_API_KEY_HERE"): Important! Replace "YOUR_API_KEY_HERE" with your actual Quandl API key.
• days_to_roll_before_expiry: Integer, e.g., 5. Specifies how many days before a contract's approximate expiration to roll. This parameter is only used if roll_method is set to "dte".
• roll_method: Determines the logic for constructing the continuous nearby series. Options:
  • "volume" (default): Uses the refined volume takeover logic.
  • "dte": Uses the days to expiration logic.
  • "oi": Uses the highest open interest logic.
• start: The first year for which to download data (e.g., 2010).
• end: The last year for which to download data (e.g., 2017).
• c_code: An index into the commodity_code vector to select the target commodity for processing (e.g., 5 for "CL" if "CL" is the 5th element in commodity_code).
• commodity_code: A vector of commodity symbols (e.g., c("C", "S", "W", "KW", "CL", "HO", "RB", "BO", "SM", "LN")). The script processes one commodity at a time, selected by c_code.
• contracts: A vector of contract month codes specific to the selected commodity (e.g., c('F', 'G', 'H', 'J', 'K', 'M', 'N', 'Q', 'U', 'V', 'X', 'Z') for Crude Oil). This must be reviewed and potentially adjusted if a different commodity is chosen via c_code.

Output

• Plot: A plot of the generated nearby series (using the selected roll_method and commodity) is displayed via ggplot2::autoplot. The plot title dynamically reflects the chosen commodity and roll method.
• CSV File: The script saves the selected nearby series to a CSV file in the same directory where the script is run. The filename is dynamic, reflecting the commodity and roll method used (e.g., CL_Nearby_volume.csv if processing Crude Oil with the volume roll method).

Dependencies

The script requires the following R packages:

• Quandl
• plyr
• tidyr
• ggplot2
• xts

You can install them using install.packages("PackageName") in your R console. The script also includes commented-out install.packages lines as a reminder. It's generally better to manage package installation outside the main operational script.