nikol
|
|
| Total Posts: 1203 |
| Joined: Jun 2005 |
| |
|
What is more correct?
I model basket of equities in multiple currencies, where base currency is, for example, USD.
A.
calculate covariance (equity, fx) -> generate future equity and fx scenarios separately and then multiply equity and fx scenarios to get equity*fx
B.
convert all equities into USD as equity*fx
calculate covariance of (equity*fx) -> model equity*fx pretending as it is traded in USD directly.
Model is more complicated, this is only high level idea. |
|
|
|
 |
Strange
|
|
| Total Posts: 1661 |
| Joined: Jun 2004 |
| |
|
Are your payoffs expressed as quanto or based on cross-performance (i.e. are your discontinuities tied to the performance in USD or in local currency)? That would probably dictate which one you'd rather do if you are trying to deal with a single product.
This said, personally I'd say it's easier to think of FX and equity vol separately, with correlation as another parameter that can be over-hedged. |
'Progress just means
bad things happen faster.’
|
|
 |
nikol
|
|
| Total Posts: 1203 |
| Joined: Jun 2005 |
| |
|
No quanto = reference within payoff is in USD and payout is in USD as well.
> easier to think of FX and equity vol separately
I agree that it is easier to think/interpret, but I have an idea that if company behind equity sells its products in USD, then EQ*LOC_USD is more natural.
it is not about hedging but rather risk reporting for b2b book. |
|
|
|
|
ronin
|
|
| Total Posts: 601 |
| Joined: May 2006 |
| |
|
> if company behind equity sells its products in USD, then EQ*LOC_USD is more natural.
You would think so, but no. Costs may be in the foreign currency. Liabilities may be in the foreign currency. USD revenue may be hedged. USD revenue may only be a proportion of the total. And so on.
> it is not about hedging but rather risk reporting for b2b book.
Then especially so. You have equity risk and currency risk. You want that to be clear in the risk report.
|
"There is a SIX am?" -- Arthur |
|
|
nikol
|
|
| Total Posts: 1203 |
| Joined: Jun 2005 |
| |
|
@ronin and @Strange
Thank you for replies.
I hear what you are saying. Translation of risk value in the management report into actual market drivers is more important then anything else.
With less emphasis - indeed, company producing locally selling to US market is sensitive to both. However, if we inject FX-scenarios into BalanceSheet model than USD sensitivity will be the biggest due to local costs (LOC) << total sales (USD). Plus, news about risk of loosing US market will have bigger impact on the company value than the risk of changing location of production costs. |
|
|
|
|
