Monte Carlo Modeling of Compounded Overnight Rates with Vasicek

Summary

The document asks whether a Vasicek short-rate model can be used to estimate compounded rates set in arrears, such as compounded SOFR. It proposes calibrating the model to observed rates, simulating paths of instantaneous rates, calculating a compounded rate along each path, and averaging the results. This outlines a Monte Carlo workflow connecting a short-rate model to an overnight-rate payoff.

The text poses the approach as a question and provides no derivation, numerical experiment, or validation. It does not address how to fit the model, represent the relevant observation and accrual calendars, or account for conventions and market pricing adjustments. Those details would be needed to judge whether the proposed average is an appropriate estimate for a particular pricing or forecasting task.

Key ideas

• A Vasicek model can be used to simulate possible paths of short-term rates.
• Compounded in-arrears rates can be calculated from simulated rates over each accrual period.
• A Monte Carlo estimate can average the compounded outcomes across simulated paths.
• The document does not establish that this procedure is correctly calibrated or suitable for market pricing.

Tags

# Modeling compounded RFRs with Vasicek


# Modeling compounded RFRs with Vasicek












I’m wondering if simple interest rates models, like Vasicek, could be successfully used for modeling compounded setting-in-arrears rates (compounded SOFR for example)?

As far as I see I can do that using monte-carlo as follows:

- Generate n paths of instantaneous rates using Vasicek fitted to real data;

- For each generated path calculate compounded rates for each date in the path;

- Take an average of calculated compounded rates.

Is that approach correct?

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.