Skip to content
All library documents

Choosing Futures Data for Monte Carlo Portfolio VaR

Article Quant Q&A · Author: Hustler885
Read the original

Summary

The document asks how to calibrate Monte Carlo simulations for value at risk on a portfolio of energy futures. The positions include gasoline and oil products such as ultra-low-sulfur diesel and West Texas Intermediate, across multiple delivery months. The specific question is whether historical prices for the same product and delivery month from the prior year should be used to model a contract held in the current year.

No answer, calibration procedure, or risk estimate is included. The question highlights a core modeling choice: matching observations by seasonal delivery month may preserve seasonal patterns, but the post does not address whether that history represents current market conditions or how to model dependencies across products and maturities. It therefore frames a useful VaR design issue without establishing that prior-year contract prices are an appropriate calibration sample.

Key ideas

  • The portfolio contains futures across energy products and delivery months.
  • The author asks whether prior-year prices for the matching delivery month are suitable for Monte Carlo calibration.
  • The document does not provide a calibration method or VaR result.
  • A complete approach would need to address seasonality, changing market conditions, and dependence across contracts.

Tags

FuturesCommoditiesRisk managementStatistics
Full text
# MonteCarlo Value At Risk for futures portfolio


# MonteCarlo Value At Risk for futures portfolio












I wanted to ask, suppose I have a portfolio of futures of gasoline and other oil products eg ULSD (Ultra Low Sulphur Diesel), WTI (West Texas Intermediate) for different months. I want to compute the MonteCarlo value at risk for the positions in the portfolio. To calibrate the MonteCarlo simulations, should I use the time series of the Future expiring in the corresponding month in the corresponding product but the year before? Eg for nov 22 ulsd should I use the nov 21 ulsd future prices to calibrate the MonteCarlo simulation ?

Thank you for the help

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.