FPI - Fractional Product Inefficiency: The Impeccable Hedge

NeoTicker indicators

Moderator: moderators

madar
rank: <50 posts
rank: <50 posts
Posts: 2
Joined: Mon Apr 05, 2010 7:20 am
Reputation: 0
Location: Singapore, SG
Gender: None specified
Contact:

Postby madar » Mon Apr 05, 2010 11:29 am

Hehe good point. I'm more interested in the intricacies of once you have $500k+ investment, what do you use for the data/hosting/broker....

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

klhh
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Fri Apr 09, 2010 7:47 am
Reputation: 0
Gender: None specified

help

Postby klhh » Wed Apr 14, 2010 6:42 am

hi, anyone pls modify this FPI indicator (attachment) by calculating the FPI using different bid and ask.



for example:
GU Bid 1.5377, Ask 1.5380
UJ Bid 93.16, Ask 93.18
GJ Bid 143.29, Ask 143.36

I would like to count GU(bid) x UJ(bid) / GJ(Ask)
1.5377*93.16/143.36=0.9992

Thank you very much.
Attachments
FPI.mq4
(1.91 KiB) Downloaded 285 times

abdullah1972
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Fri Jun 11, 2010 3:21 am
Reputation: 0
Gender: None specified

Postby abdullah1972 » Sat Jun 12, 2010 8:06 am

i read in this topic that we can make the speared 0 that we can sell from bid and buy from ask how i can do this in MT4?
thankx micale and everyone how post useful idea or indi

rklee1
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Sun Oct 03, 2010 12:59 am
Reputation: 0
Gender: None specified

Postby rklee1 » Sun Oct 03, 2010 5:26 am

I've downloaded the file. How do i open it to look at all the charts you are showing on the original post?

hydefus
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Fri Mar 11, 2011 12:31 pm
Reputation: 0
Gender: Male

Postby hydefus » Tue Mar 15, 2011 1:11 am

Instead of checking for the FPI indicator, i think a better idea is to build an EA that monitor the net profit for all the three orders in the ring, let's say every 100 milliseconds, and check if we have a positive amount. FPI indicator could be useful to know how much we could earn based on the history of the ring.

The problem with all those techniques, even what you propose is that they conditioned by the spread. In the example you show, you put very small spreads if you compare with the reality today.

Best wishes!

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.

williamson
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Wed Mar 16, 2011 12:40 pm
Reputation: 0
Gender: Male

Language Translator

Postby williamson » Wed Mar 16, 2011 12:51 pm

Hello Michal,
I?m sad to see that you have spent so much time in this FPI. I wish we could have met sooner. In 2001 I designed a very interesting triangular arbitrage system in excel. I got outside quotes and placed the orders by phone. My trades had at least 5 or 6 legs and at the end, if I end up positive I was happy. Sometimes I had to wait couple of seconds for it to turn positive and place the closing leg. I turned $150,000 into more than a $1,000,000 in less than 2 years. I was 27 years old, spent more than $200,000 in Las Vegas, had a beautiful boat, Mercedes and bachelor pad to die for. I was doing this until 2003 when my bucket shop broker ran away with a lot of my money.
I took a break, then opened an account with Refco and they were gone too. I decided to learn some other things in life, until Triangular Arbitrage came to mind again in 2008. The only thing was that I didn?t want to think, calculate manually or place trades by hand. So I created what is literally a money machine.
It calculates every Triangular Arbitrage available in the world real-time and places the orders that are only profitable. With some slight adjustments it can do the same thing for FPI (although I am not a firm believer of FPI). Slippage is horrible but manageable. A very interesting fact is that in all the complexities, thousands of combinations (rings) and millions of calculations per second, my algorithm is so simple that it can run in any laptop with XP and Excel 2003. The largest number of combinations (rings) placed I have seen totally executed and confirmed in a second were 4 (12 separate trades). I love Currenex and the Fix Gateway!
The worst thing about this is that no brokerage house wants or can?t deal with me and I have nowhere to hook my system.
Again, we should have met before; I would have saved you a lot of headaches.
Thank you for reading such a long post but I deserve it, I found your thread last night I didn?t even skipped a page.
If you have any idea, let me know![quote]

I was so excited to find this lens.
Introduce to us language translator

Capella
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Wed Mar 30, 2011 6:50 pm
Reputation: 0
Gender: None specified

My way of using Triangular Hedging

Postby Capella » Sat Apr 02, 2011 10:28 am

Hi!

Is this thread dead or?

Anyway, I found this thread some days ago. And although the topic as such is not new to me, it gave me little inspiration to take a closer look at this strategy again, and want to express some ideas for its practical use. Maybe I can give some of you some inspiration as well.

I've used hedging techniques when scalping between brokers with different results. Very few brokers allowes this to be done on their real servers. One such attempt was done on a demo-server with a broker (I don't want to reveal its name here), where I used other brokers as seeders and this broker as feeder of data, and then made arbitrage hedging between brokers, and was trading with a fast scalping with one of the brokers. You can see the result from one weeks trading below, which shows an amazing result. Remember though, that it was on a demo account (I've removed my name, account number and broker name, all other data are intact). I initially let the robot use 2.5 lot, then changed it back to 1 fix lot-size. The last orders where closed manually. And I stopped the robot after one week, since it generated so much data, and I knew that it was working under these conditions. The htm-file is almost 9 Mb large so I've zip-compressed it to a 370 Kb. It can be downloaded from...

ArbitrageHedgeScalperDemoResult

I'm fully convinced that using FPI alone for us ordinary traders is useless. Spread and other costs, together with time delays causes the possible profit to be lost. Scalping on tic-basis (opening and closing positions within seconds as I showed above) is also not possible with the majority of brokers. Beleieve me, I've tried! And those who allowes it only allow it on their demo-servers.

My suggested strategy is to trade with a ring of hedged currencies, using FPI as one of several indicators to enter and exit trades, and not use FPI alone.

If we take one such ring as an example... EURUSD + GBPUSD + EURGBP.
For this ring the two different trading options are BUY + SELL + SELL or SELL + BUY + BUY (since each one of the three currencies must be hedged BUY-SELL).

Now, use your favorite indicators to find trends for each one of the three currency pairs. It will be hard, if not impossible, to find a trend for all three in the correct direction at the same time, so you look for times when two currency pairs are moving in the right trend direction. For the above currency pairs it must be BUY + SELL + 0 or BUY + 0 + SELL or 0 + SELL + SELL or SELL + BUY + 0 or SELL + 0 + BUY or 0 + BUY + BUY, where 0 indicates no trend. Then open two positions for the two currency pairs that you have trend for. This will hedge one of the three currencies, but not all three. After a while you will see that the trend changes, and you take action accordingly. That could be to close on of the opened positions, to close both of them, to close one and open a third (for the currency pair that was not in trend when you opened the first two positions, but which now are in trend).

Using this technique the goal is to hedge all three currencies in the ring, but you don't do it immediately. That is, you don't open and close them all at the same time. You start hedging one currency by opening two positions, then wait for trend changes to open and/or close positions, so you will eventually hedge two currencies, and the one or the two currencies of the three that are hedged will change within the ring. It works well for manual trading, but is best suited for automatized trading.

You can trade on any time-frame. Using automatized systems I recommend trading on M1, and to use very small trailing stops to scalp a few pips from each trade.

Good luck!

pika
rank: 50+ posts
rank: 50+ posts
Posts: 74
Joined: Tue Sep 21, 2010 5:00 pm
Reputation: 0
Gender: None specified

Postby pika » Sun Apr 03, 2011 1:36 am

Ultimately you need a trend in one or more pairs to move long enough in order to mitigate the loss from the corresponding hedging pairs. The spread incurred on all the pairs makes such a strategy rather costly. In exchange, you are just buying time hoping for a strong trend to emerge in at least one of the pairs in your favour. It works fine if this happens, but the losses will be huge if the trend turn against you before you can realise your profits. Remember, you will not win all the pairs, at least one of them will be a loss and you need to cover it with enough profits from another winning pair. Hence, I don't recommend this strategy. However, it may be more useful with this strategy when all the ring pairs are non-trending and waiting for a breakout to a new direction instead.

thunderball
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Wed Feb 09, 2011 11:16 pm
Reputation: 0
Gender: Male

Re: FPI - Fractional Product Inefficiency: The Impeccable He

Postby thunderball » Tue Jun 21, 2011 6:28 pm

Hi, Michal,

Maybe these questions was already asked, but:

"to predict the behavior of one of the FX symbols in ring" - have you any ideas HOW that can be implemented? HOW determinate which pair/currency disbalance in ring? HOW determinate the level of influence of each pair in ring on integrity/balance of ring?

Thank you for answers.

mabdullah
rank: <50 posts
rank: <50 posts
Posts: 1
Joined: Sun Aug 14, 2011 4:43 pm
Reputation: 0
Gender: None specified

Re: Language Translator

Postby mabdullah » Sun Aug 14, 2011 7:17 pm

Hi, can i have your email plz so we could have more insight into this arbitrage algo? or you could drop an email at mubashir.abdullah@yahoo.com.

many thanks, and i look forward to hear from you

regards
mubashir

williamson wrote:Hello Michal,
I?m sad to see that you have spent so much time in this FPI. I wish we could have met sooner. In 2001 I designed a very interesting triangular arbitrage system in excel. I got outside quotes and placed the orders by phone. My trades had at least 5 or 6 legs and at the end, if I end up positive I was happy. Sometimes I had to wait couple of seconds for it to turn positive and place the closing leg. I turned $150,000 into more than a $1,000,000 in less than 2 years. I was 27 years old, spent more than $200,000 in Las Vegas, had a beautiful boat, Mercedes and bachelor pad to die for. I was doing this until 2003 when my bucket shop broker ran away with a lot of my money.
I took a break, then opened an account with Refco and they were gone too. I decided to learn some other things in life, until Triangular Arbitrage came to mind again in 2008. The only thing was that I didn?t want to think, calculate manually or place trades by hand. So I created what is literally a money machine.
It calculates every Triangular Arbitrage available in the world real-time and places the orders that are only profitable. With some slight adjustments it can do the same thing for FPI (although I am not a firm believer of FPI). Slippage is horrible but manageable. A very interesting fact is that in all the complexities, thousands of combinations (rings) and millions of calculations per second, my algorithm is so simple that it can run in any laptop with XP and Excel 2003. The largest number of combinations (rings) placed I have seen totally executed and confirmed in a second were 4 (12 separate trades). I love Currenex and the Fix Gateway!
The worst thing about this is that no brokerage house wants or can?t deal with me and I have nowhere to hook my system.
Again, we should have met before; I would have saved you a lot of headaches.
Thank you for reading such a long post but I deserve it, I found your thread last night I didn?t even skipped a page.
If you have any idea, let me know!

I was so excited to find this lens.

Please add www.kreslik.com to your ad blocker white list.
Thank you for your support.


Return to “NeoTicker indicators”