My question is regarding the European Commission regulation on standardizing the information in the key information documents for PRIIPs. In the Annex IV of the regulation, one can find the explanation on the determination of an MRM class for the category 3 products (paras. 16-24)

Now my question is regarding calculation of the return of the underlying contract, as explained in para. 22, and specifically the formula in 22c:

$Return = E[Return_{risk-neutral}] - E[Return_{measured}] - 0.5\sigma^2N - \rho\sigma\sigma_{ccy}N$

Could anybody give more details on this formula? I personally do not think I fully understand the first two terms in the formula.

I also tried to reproduce the category 3 calculation as given in the Commission presentation, but I miss the point after summing up the simulated daily returns, i.e. step 9.

Thank you!


Actually, it is not very clear the legislation. However, from some slides that EIOPA used in a conference I tried to build back their computation and what I found is that:

1) you simulate 10k with lenght T, so you have a matrix 10k x T

2) you sum over the Ts so you get a vector 10k x 1

3) you have to go in the risk neutral world so from every element from 1 to 10k you remove the historical mean, add the risk free and remove FX risk/variance

  • $\begingroup$ Thank you. I also tried to reproduce the calculation (I guess from the same presentation[1]), however it is not possible to get the exact same result. Their final return is equal to 1.057013249, which is 0.055447241 as a lognormal value. However, Sum of Returns = 0.029342; -10*Mean = 0.022725205; -10*Variance/2 = -0.000667217; With a risk free rate of 1.2% p.a., i don't see how to end up with 0.055447241. How did you solve that? [1] ec.europa.eu/info/system/files/risk-section-kid-11072016_en.pdf $\endgroup$ – Boris May 16 '17 at 10:20

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.