FPI - Fractional Product Inefficiency: The Impeccable Hedge
Moderator: moderators
- TheEconomist
- rank: <50 posts
- Posts: 22
- Joined: Thu Nov 02, 2006 7:13 pm
- Reputation: 0
- Gender:
Just a side thought -- perhaps it may work if we reverse engineer the swaps and convert them to interest rate differentials.. So the script would choose from a set of stable rings and filter out the "inflated" rings with large, but unstable payouts, where the forex rate weight in total payout is greater... but could it be that, in this case, good rings to be turned negative? or just applying the interest rate differential criteria we'd get a hedge on the swaps themselves?
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.
- michal.kreslik
- rank: 1000+ posts
- Posts: 1048
- Joined: Sat May 13, 2006 2:40 am
- Reputation: 36
- Location: Monte Carlo, Monaco
- Real name: Michal Kreslik
- Gender:
TheEconomist wrote:Just a side thought -- perhaps it may work if we reverse engineer the swaps and convert them to interest rate differentials..
If you mean doing the calculation based on the central banks' interest rates, then this wouldn't work out. Well unless you got access to the central bank's repo interest rate.
In the FPI ring's composite interest rate calculation, one can only include the publicly available rollover rates published by brokers or other institutions where one can really get these rates.
Michal
-
- rank: <50 posts
- Posts: 8
- Joined: Wed Nov 22, 2006 7:45 pm
- Reputation: 0
- Location: San Diego, CA
- Gender:
Data
Hi all,
I wrote an EA to give me an idea of how this will work on various brokers. Results vary dramatically, but we already knew that, didn't we?
It self-intializes for the trading pairs at a given broker and then populates the display with the available rings, then tracks them for an entry signal. It shows one pair currently trading, but I forced that entry point in order to track data without waiting for a "good" signal.
I ran it overnight and it showed the following results in the ring data block:
1, 2: Ring name followed by net swap based on one lot for pair 1 & 3, with variable lot size on #2.
3, 4, 5: Fraction current/high/low during this EA session.
6, 7, 8: Projected current/high/low "balance" based on a stored standard price.
9: Net of high/low price.
10: Net of high/low price AFTER "regular" spread is deducted for each pair in the ring. NOTE - this does not consider spread fluctuations.
11: Combined spread one would pay for the three currencies to enter a ring.
The column to watch is the second to last. If it has a relatively large positive value, that ring has possibilities.
Have fun,
Bill
I wrote an EA to give me an idea of how this will work on various brokers. Results vary dramatically, but we already knew that, didn't we?
It self-intializes for the trading pairs at a given broker and then populates the display with the available rings, then tracks them for an entry signal. It shows one pair currently trading, but I forced that entry point in order to track data without waiting for a "good" signal.
I ran it overnight and it showed the following results in the ring data block:
1, 2: Ring name followed by net swap based on one lot for pair 1 & 3, with variable lot size on #2.
3, 4, 5: Fraction current/high/low during this EA session.
6, 7, 8: Projected current/high/low "balance" based on a stored standard price.
9: Net of high/low price.
10: Net of high/low price AFTER "regular" spread is deducted for each pair in the ring. NOTE - this does not consider spread fluctuations.
11: Combined spread one would pay for the three currencies to enter a ring.
The column to watch is the second to last. If it has a relatively large positive value, that ring has possibilities.
Have fun,
Bill
- Attachments
-
- neuimex ring.gif (40.11 KiB) Viewed 4985 times
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.
- eagles
- rank: <50 posts
- Posts: 28
- Joined: Mon May 22, 2006 6:38 pm
- Reputation: 0
- Location: Plano, TX
- Gender:
Bill,
There are 2 rings in your screenshot that have positive swap:
USDCHF_CHFJPY_USDJPY_SSL
AUDUSD_USDJPY_AUDJPY_LLS
Do you see any potential in these rings as a carry trades?
Also, in your lot size display it shows 1.22/1.00/1.00, but your description above says "one lot for pair 1 & 3, with variable lot size on #2" --- Is this just the way you set up the display, or are you trying different variations?
Great work though, as it puts some definitive information about 'real' rings.
Regards,
Jack
There are 2 rings in your screenshot that have positive swap:
USDCHF_CHFJPY_USDJPY_SSL
AUDUSD_USDJPY_AUDJPY_LLS
Do you see any potential in these rings as a carry trades?
Also, in your lot size display it shows 1.22/1.00/1.00, but your description above says "one lot for pair 1 & 3, with variable lot size on #2" --- Is this just the way you set up the display, or are you trying different variations?
Great work though, as it puts some definitive information about 'real' rings.
Regards,
Jack
-
- rank: <50 posts
- Posts: 8
- Joined: Wed Nov 22, 2006 7:45 pm
- Reputation: 0
- Location: San Diego, CA
- Gender:
Hi,
Yes, I used the screen shot from this broker because it had postive swap, and I had previously posted that I coudn't find any. With some brokers there are no positive swap rings.
I prefer certain correlation trades for positive swap. The return is greater, I think.
Yes, the order data shows 1.22 first. They are simply out of order. I didn't take time to clean up that part of the code as it has no impact on my results.
Take care,
Bill
Yes, I used the screen shot from this broker because it had postive swap, and I had previously posted that I coudn't find any. With some brokers there are no positive swap rings.
I prefer certain correlation trades for positive swap. The return is greater, I think.
Yes, the order data shows 1.22 first. They are simply out of order. I didn't take time to clean up that part of the code as it has no impact on my results.
Take care,
Bill
-
- rank: <50 posts
- Posts: 8
- Joined: Wed Nov 22, 2006 7:45 pm
- Reputation: 0
- Location: San Diego, CA
- Gender:
how do I download/get a copy of your EA? I would like to test your EA.
Hi,
Sorry, I'm not distributing this EA. It is a rough program I wrote to test the viability of rings in the "real world" where spreads are an issue, and which dramatically reduce the opportunities for this trade.
I won't be cleaning up the code unless I find a broker where the trades are not only possible, but frequent.
But, I thought that the results were be interesting enough to share. I hope it has been of some help to this group.
Take care,
Bill
Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.
Thank you for your support.