3
$\begingroup$

Under the FRTB we do an downwards and upwards shocks reating to the each of the risk factors.

FX risk is also calculated as the change in respect to the reporting currency to each of legs of the trade.

For example a USDJPY FX option with say $10m Notional, when the reporting currency is EUR, we need to calculate the FX risk based on EUR to USD and EUR to JPY. This would be;

Notional x delta = USD FX position

Notional x strike x delta = JPY FX position

These positions are used to calculate the FX risk.

When calculating the curvature - the FX shocks changes the delta, and we get new positons in USD and JPY. Quite streight forward.

However if I shock the USD and JPY interest rates, I get a change in the OV giving the GIRR CRV risk impact. These GIRR risk give rise to translation risk as well in that we have a risk position in USD and JPY. This I have modelled.

However the shock in interest rates leads to a change in the delta, which in tern leads to a change in the FX position which is cross gamma. Should this cross gamma impact be picked up in the CRV calculations as well?

Improve this question
$\endgroup$
3
  • $\begingroup$ marketsrisks.com/2016/08/24/curvature-risk-charge :( $\endgroup$
    – Dimitri Vulis
    Sep 4 at 23:57
  • $\begingroup$ I saw that - just general stuff - does not address the issue above... $\endgroup$
    – Ben Watson
    Sep 5 at 0:14
  • $\begingroup$ I dont know the answer as applied to the regulation, but the cross gamma effect is applicable to any product (indeed those more simple), e.g. a received position in a USD IRS when the accounting currency is EUR. Do these cross-gamma effects get applied for these products also, i.e . is the entire framework consistent (or inconsistent) $\endgroup$
    – Attack68 ♦
    Sep 7 at 16:46

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.