I have found a very useful post regarding the use of Monte Carlo simulaton to obtain portfolio Value at risk, based on Cholesky decomposition, random variates, etc. This post I'm talking about is: Is there a step-by-step guide for calculating portfolio VaR using monte carlo simulations

However, I don't understand if those same steps can be followed if I have different asset classes in my portfolio. For example, what if I have an Apple stock, a US Treasury Bond and a derivative (maybe an option on that same stock) with weights being 35%, 45%, 20% respectively?? Should I use duration, delta, gamma, etc?? How would this work? Thanks a lot.

  • $\begingroup$ I'm stuck in the same place. Are we supposed to do the MC for all the factors of all the assets in my portfolio? For example, I'll take each historical stock returns as the equity factor, spot rates at different key points(like 1Y, 2Y, 3Y, 5Y, etc.) as the FI factor, and of course if I have various types of FI, I should have every key point historical rate series. As to the option, I would use the underlying's return series and rf as two more factors. Next will be just following the steps in the link, covariance, Cholesky, and so on. I don't think treating FI as equity is appropriate. $\endgroup$
    – Aria
    Feb 8 '21 at 6:28
  • 1
    $\begingroup$ The comment above was edited to fit in a comment and converted from answer. $\endgroup$
    – Bob Jansen
    Feb 8 '21 at 16:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.