Menu

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Basket Knights - A Basket Trading Round Table 1,311 replies

BASKET TRADE Management: Using multiple trade management EA's on basket trades 9 replies

TMS basket: Dashboard for a basket of MA-Filtered RSI Signals 135 replies

Carry/Basket Trade Idea 17 replies

euro based carry trade basket 7 replies

Attachments: Currency Basket and Carry Trading
Exit Attachments

Currency Basket and Carry Trading

Quoting firehorse
Did you have a system for picking those two particular pairs as opposed to any other pairs.
No. I read about these two pairs on another forum. I started demoing to check on the interest returns, but soon realised there were profit taking opportunities.


Quoting firehorse
Also, do you have a system for changing the pairs if circumstance change?
In my opinion it would be a concern if USDJPY is near historical high. I would need to scale down exposure, or change pairs. Consideration would probably be corelation, trend, stability of the pairs.


Regards,
Matsu
  • Post #42
  • Quote
  • Edited Nov 25, 2005 9:42am Nov 19, 2005 2:29pm | Edited Nov 25, 2005 9:42am
Quoting BobMM
Ok, here's a comparison of basket prices over 5 years. Included is: mine, ElectronicSavants, and DGCFX's. Those last two have a very similiar basket with a slightly different weighting. Both of them went into negative NAV territory in the start of 2002 at their default settings because they use up 50% of the account as margin when doing 15:1 leverage. (exotics and hedges)

If you guys find flaws with what I'm doing, point em out! I'm as amazed and impressed by this basket's performance as you may be, and I want to make sure I can make the most profitable system from it possible.

-Aksel
Thanks Aksel, I changed my basket pricing backtesting strategy to use cumulative percentage difference for each day, instead of pip price. I am testing ES's basket so I multiplied the daily percentage difference based upon the percentage weighting factor for that currency in USD. I sum up the total percentage each day and have a cumulative running total of basket price % change. The DD has gone as low as 4.4%. Any suggestions? Is there something I am doing wrong or should be adding to the formula?
SJ
Quoting sjburdi
Thanks Aksel, I changed my basket pricing backtesting strategy to use cumulative percentage difference for each day, instead of pip price. I am testing ES's basket so I multiplied the daily percentage difference based upon the percentage weighting factor for that currency in USD. I sum up the total percentage each day and have a cumulative running total of basket price % change. The DD has gone as low as 4.4%. Any suggestions? Is there something I am doing wrong or should be adding to the formula?
DD should be much larger than that, 2/20/04-5/11/04 shows up around -6% DD for me. From the way you phrased your methodology, it's hard to figure out where our methods diverge. I'll explain mine in more detail and if you do the same we can find where we differ.

I have a huge table of prices... so if AUDJPY is in the basket, I first make a new column of the USD equivalent of AUDJPY. (AUDJPY*1/USDJPY). This is the column I use for basket price determination. On the day it is brought in I set another column with the total change of AUDJPY USD equiv based on today's price change. So for say, 3 days where the price went up 10% each day, we'd see:

0%
10%
10%
10%

Final basket price is Weighting*%daily change of each pair + next pair, etc and denominated in dollars the whole time. Maybe the difference is in how we did our weightings. Looking off his exposure tab, I did the weighting as 11364/85512*(%chg) + 6818/85512(%chg), so I think I used the units of each pair/total # of units. This is a real head wrapper now, because those are units of the currencies not translated into USD, so did I make a mistake here? But the units shouldnt matter, since the final currencies are expressed in USD, right?

I'm thinking I did it correctly, since to decide how many of each to buy you start with his amount of NZDUSD, and then use ratios to buy units of the others, and it's in a set amount. So as long as you have the right ratio of units you should duplicate his performance methinks. Thoughts?

I suspect we did our weightings differently and that's the reason for the discrepancy in our findings. Anyone else want to chime in and offer their opinion? How do we do his weighting?

-Aksel
Quoting BobMM
DD should be much larger than that, 2/20/04-5/11/04 shows up around -6% DD for me. From the way you phrased your methodology, it's hard to figure out where our methods diverge. I'll explain mine in more detail and if you do the same we can find where we differ.

I have a huge table of prices... so if AUDJPY is in the basket, I first make a new column of the USD equivalent of AUDJPY. (AUDJPY*1/USDJPY). This is the column I use for basket price determination. On the day it is brought in I set another column with the total change of AUDJPY USD equiv based on today's price change. So for say, 3 days where the price went up 10% each day, we'd see:

0%
10%
10%
10%

Final basket price is Weighting*%daily change of each pair + next pair, etc and denominated in dollars the whole time. Maybe the difference is in how we did our weightings. Looking off his exposure tab, I did the weighting as 11364/85512*(%chg) + 6818/85512(%chg), so I think I used the units of each pair/total # of units. This is a real head wrapper now, because those are units of the currencies not translated into USD, so did I make a mistake here? But the units shouldnt matter, since the final currencies are expressed in USD, right?

I'm thinking I did it correctly, since to decide how many of each to buy you start with his amount of NZDUSD, and then use ratios to buy units of the others, and it's in a set amount. So as long as you have the right ratio of units you should duplicate his performance methinks. Thoughts?

I suspect we did our weightings differently and that's the reason for the discrepancy in our findings. Anyone else want to chime in and offer their opinion? How do we do his weighting?

-Aksel
Aksel: You are right. I am doing the weighting totally differently. I do not use Units to determine weighting. What I do is determine the dollar amount invested for each pair. Then I determine what percentage of the total amount invested that pair represents. So now I have my weighting in todays dollars. Then i have to go back in time to determine my percent weightings 5 years ago - in order to get an accurate starting point based upon prices 5 years ago. To do this, I determine the price percentage difference for each pair from 5 years ago to todays price. Now I apply that percentage change to each pairs current today's price and determine a new 5-year ago weighting for each pair and start from there. Then I get a percentage difference each day for each pair and total up the basket. This strategy may be flawed as well, as I continue to think about it. Please let me know your thoughts here.
SJ
Quoting sjburdi
Aksel: You are right. I am doing the weighting totally differently. I do not use Units to determine weighting. What I do is determine the dollar amount invested for each pair. Then I determine what percentage of the total amount invested that pair represents. So now I have my weighting in todays dollars. Then i have to go back in time to determine my percent weightings 5 years ago - in order to get an accurate starting point based upon prices 5 years ago. To do this, I determine the price percentage difference for each pair from 5 years ago to todays price. Now I apply that percentage change to each pairs current today's price and determine a new 5-year ago weighting for each pair and start from there. Then I get a percentage difference each day for each pair and total up the basket. This strategy may be flawed as well, as I continue to think about it. Please let me know your thoughts here.
So clear me up on this one point: Are you dollar-weighting it based on the prices 5 years ago or prices today? Because, prices change. I'm not quite sure which method is right, quite frankly. And you're not changing the weightings, right? Or are you adjusting the weighting daily based on dollar fluctuations?

Better yet, how many of each unit do you hold today? I'm thinking dollar-basing isn't exactly following his model, because he mentioned he didn't expect changes to weighting except very rarely, if at all. The point is to backtest his system as it is now, with the weightings he and dfgx came to in hindsight, right? Rather than what it may have been 5 years ago?

-Aksel
Quoting BobMM
So clear me up on this one point: Are you dollar-weighting it based on the prices 5 years ago or prices today? Because, prices change. I'm not quite sure which method is right, quite frankly. And you're not changing the weightings, right? Or are you adjusting the weighting daily based on dollar fluctuations?

Better yet, how many of each unit do you hold today? I'm thinking dollar-basing isn't exactly following his model, because he mentioned he didn't expect changes to weighting except very rarely, if at all. The point is to backtest his system as it is now, with the weightings he and dfgx came to in hindsight, right? Rather than what it may have been 5 years ago?

-Aksel
Interestingly, I have found that the weightings really do no matter a whole lot. If you invested a certain amount and divided it equally among the 10 pair or if you take dfgx's weightings and then calculated what percentage that equals in each currency based upon price 5 years ago, the difference is not that great to total basket price percentage change.
SJ
  • Post #47
  • Quote
  • Edited 2:42am Dec 2, 2005 12:55am | Edited 2:42am
I'm doing it with 33 pairs and find weighting and exposure don't effect my decisions.

I just brought in slowly to what I considered a good mix of yielders and slowly built up on the winners from there. I now have a basket and I play with that. Nothing more, nothing less.

I've also started buying box options to hedge big news nights. But I've found it's really not nessesary due to the balance of the basket. For example last weeks 100 points runs changed nothing. I woke up to see my unrealized basicly the same value. I would have been none the wiser to the news if I had not looked at the charts. It was a bit of a shock and I had to look more than twice!

So far I'm sitting on 14% for the month with only ~23% of equity tied up.
grist for the mill
  • Post #48
  • Quote
  • Edited 3:00am Dec 2, 2005 2:52am | Edited 3:00am
...and this is all thanks to Electric and his posts

My style is to have the negative weights to counter big moves which still earn some Interest.

Have the Earners where I balance them around zero (not sure how to term this)

And my Trenders where I get net short/long and play the breaks.

Slightly more time consuming but it gives me lots of time to think. It does cause some head aches and I sometimes can not be bothered thinking about it to be honest but it's nice to have time on my side.

I'm also looking to pin point big fundamental news and get flat or square and ride out the change.

Fun for the whole family really.
grist for the mill
Quoting JackJones
...and this is all thanks to Electric and his posts

My style is to have the negative weights to counter big moves which still earn some Interest.

Have the Earners where I balance them around zero (not sure how to term this)

And my Trenders where I get net short/long and play the breaks.

Slightly more time consuming but it gives me lots of time to think. It does cause some head aches and I sometimes can not be bothered thinking about it to be honest but it's nice to have time on my side.

I'm also looking to pin point big fundamental news and get flat or square and ride out the change.

Fun for the whole family really.
Agreed: I have to thank Electric for introducing this system to FF. What a run the last 45 days has been for C&C standard. I am up over 70% in UPL. What goes up 70% can also go down 70%. That is the lesson I take from the recent run.
SJ
Quoting ElectricSavant
Your assuming your exposure is not the culprit and WHEN you you "enter", "scale in" or "add to"...
Its Buy and Hold since 10/9/05 and up over 70% on UPL
SJ
  • Post #51
  • Quote
  • Edited 9:49am Dec 8, 2005 9:47am | Edited 9:49am
I've just began a demo account on Oanda. Starting balance £998 (had to minus away the 100k they gave me to start).

Decided to go:

long GBP/JPY 2000 UNITS

short EUR/JPY 2000 UNITS

From what i see, these pairs are highly correlated, so i should be nicely hedged. Im interest positive overall from being long GBP. Any thoughts?

*edit* leverage is 20:1, marging used on trade is £167.81.
  • Post #52
  • Quote
  • Last Post: Jan 18, 2007 9:40pm Jan 18, 2007 9:40pm
Whatever happened to basket trading? Why are all of ElectricSavants posts edited?

I'm looking for a good hedge strategy to earn interest on.
0 traders viewing now