Quote from oldtime:
right, but I'm very rarely 1:1:1:1, so I attempt to buy or add to dips and rallies in each pair independently.
what would the ratio be if I was long unequal amounts of aud and gbp?
I like the idea though. I'll have to think it through
if I'm long gbp and aud and go short eur.usd, aren"t I just cancelling out my short usd?
For unequal exposures you'd just have to work out your net exposure for a single currency in that single currency's units and compare it to the possible substitute (like your EUR half for more USD, or the other way around,) knowing you won't get a perfect fit.
Now, I'm tired and I'm going to write this out without really proofing it..so excuse any math errors in advance, but let's toss up some random position sizes for the sake of example:
(numbers below assume exchange rates ~ closing levels today)
Long 1.25 lots AUD/USD -- $125,000AUD by $(129,625)
Long 0.75 lots GBP/USD -- £75,000 by $(121,252)
Short 1 lots EUR/AUD -- â¬(100,000) by $127,410AUD
Short 1.2 lots EUR/GBP -- â¬(120,000) by £98,100
Netting you and inventory of:
Owing
$(250,877) USD
â¬(220,000)
Holding
$252,410 AUD
£173,100
Now, without reducing any exposure to your AUD and GBP holdings (since you're still rather bullish on them,) you could shift around your USD and EUR exposure a bit depending on how you want it weighted:
Say we want to remove as much USD risk as possible (and shift that risk to the Euro, you don't remove the risk itself since you gotta borrow something to hold your AUD and GBP.)
Short 1.9 lots of EUR/USD, which adds â¬190,000 additional Euros to the â¬220,000 liability but also puts a positive position of $251,142 USD into your exposure/holdings which effectively brings your total USD exposure to near 0.. (well, you'd be a few hundred net positive USD but let's not split hairs when working out an example like this

)
Owing
â¬(220,000) + â¬(190,000) == â¬(410,000)
Holding
$(250,877) + $251,142 == $265 USD
$252,410 AUD
£173,100
You could do the reverse and try to remove as much EUR as possible as well, piling the risk into USD instead... but the imbalance wouldn't be nearly as close to zero as the best you could do (without adding additional exposure beyond your basket) leaves ~â¬30k hanging.
----
Now, the problems with this:
- Additional cost to put on the additional trades (more spread paid, more commissions paid,) so it better be something you're convinced about or a risk you want to mitigate.
- Additional swap / cost of carry interest charges.
- Psychologically you gotta keep your mind focused on the basket's total P/L, not the individual trade's P/L... you are playing a game of net exposure here and sweating over a loss on one component of the basket won't do you any good (nor will cheering for one component's win.)
- If you are planning on doing this shift for most of your basket's time horizon, it clearly would have been cheaper and more sane to just not include the undesired currency in the first place.. so this is really just for short term adjustments and for short term event risks.. not so much a core play within your basket.
Lastly, to do your own calcs: Identify the base and quote currency of your target pair, plus the exchange rate, this will give you your exposure to the currencies involved.. then just add long or short exposures as you deem required to even things out.
Fun, no?