FPI - Fractional Product Inefficiency: The Impeccable Hedge

NeoTicker indicators

Moderator: moderators

User avatar
michal.kreslik
rank: 1000+ posts
rank: 1000+ posts
Posts: 1032
Joined: Sat May 13, 2006 2:40 am
Reputation: 0
Location: Monte Carlo, Monaco
Real name: Michal Kreslik
Gender: Male
Contact:

Postby michal.kreslik » Fri Dec 28, 2007 11:47 pm

Annu wrote:Michal: just out of curiosity, what is your approximate average holding duration of such a FPI-position? (seconds, minutes, hours)?

Regards,
A.


I have not calculated the averages, but it ranges from a couple of minutes to many hours. I'm working on testing the automated FPI framework now and the more data I collect, the more statistically correct the testing results will be.

Michal

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

User avatar
michal.kreslik
rank: 1000+ posts
rank: 1000+ posts
Posts: 1032
Joined: Sat May 13, 2006 2:40 am
Reputation: 0
Location: Monte Carlo, Monaco
Real name: Michal Kreslik
Gender: Male
Contact:

Postby michal.kreslik » Fri Dec 28, 2007 11:59 pm

daniil wrote:Michal! it seems you use Limits to fill, aren't you?


With the FPI arbitrage, the best price is the current price. Point blank. You can get a better price (making the FPI value more extreme), but the problem is that you don't know whether the better price is a price above the current price or a price below the current price for the particular pair because at the time you get the presumably better price, the other pairs in an FPI ring might have moved to another prices as well and you would have to recalculate the FPI value again.

That's why the best combo is a market order with a razor sharp execution.

You have to think outside the box with the FPI concept :) Limit is not necessarily "this or better price" in this case.

daniil wrote:do you use DEMO for testing?


The data collected from the demo and live accounts at EFX are the same as I've found out. I'm testing my FPI framework on the collected data and on the demo account.

Michal

davidf
rank: <50 posts
rank: <50 posts
Posts: 24
Joined: Fri May 19, 2006 6:10 pm
Reputation: 0
Gender: Male
Contact:

Lot size again

Postby davidf » Tue Jan 01, 2008 10:45 am

I tryed calcutale lot size, but I dont know whether it is correctly.
bid ask
EUR/USD 1,4637 1,4640
USD/CHF 1,1304 1,1309
EUR/CHF 1,6553 1,6558

SSB ring: paid get
sell EUR 10000,00 14637,00 USD
sell USD 14637,00 16545,66 CHF
buy CHF 16545,66 9992,55 EUR

for

BBS ring: get paid
buy EUR 6830,60 10000,00 USD
sell CHF 11306,69 6830,60 EUR
buy USD 9997,96 11306,69 CHF

I get lower EUR or USD value that I paid. I think it is because calculated EUR/CHF price is lower that market price. Calculate I lot size correctly?

DavidF

obx
rank: <50 posts
rank: <50 posts
Posts: 41
Joined: Fri May 26, 2006 12:35 pm
Reputation: 0
Gender: Male

Postby obx » Thu Jan 03, 2008 5:24 pm

Michal,

What is the longest bar time frame / tick count you use for FPI pairs?

Thanks,

obx

cc
rank: <50 posts
rank: <50 posts
Posts: 2
Joined: Wed Dec 26, 2007 7:10 pm
Reputation: 0
Gender: None specified

Postby cc » Thu Jan 03, 2008 7:07 pm

Michal,
How do you find the MB/EFX API? The Navigator seems quite unstable.
I wrote a SDK API program to monitor a single FPI ring but the navigator instance seems to crash after about a day, hardly the stuff to which you would want to entrust other peoples money.
Cheers
Craig

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

gustavodemelo
rank: <50 posts
rank: <50 posts
Posts: 4
Joined: Wed Nov 14, 2007 3:01 am
Reputation: 0
Gender: Male
Contact:

Postby gustavodemelo » Thu Jan 03, 2008 8:01 pm

Michal, testing the FPI and that results, sometimes I found diferent values (sum of profit) then the others, using the same FPI to open the same rings and to close the same rings. Is it showing some " ineficiency" in the FPI?

I change this factor and now I'm using just the sum of values (product no more), and the results are more corrects comparing the sum of profit with the factors.

The factor is very simple. I create using the sum of market values (+) when is BID and (-) when is ASK. When the pair have JPY currency, I divide by 100 before sum. I'm start using now this factor (FSI) in my application runing in Meta Trade 4 and works well.

It can open and close the orders but I need to create an application (front end) to control the CSV files and pther parameters to send to the EA.

My questions:
1) Do you find this diferences using FPI?
2) Do you think to use other math logic to generate a factor?

I'm convinced that the most important step in Hedge is:
When exately to open and close the rings. And I tested that FPI some time fails to show this entry.

Regards.

User avatar
michal.kreslik
rank: 1000+ posts
rank: 1000+ posts
Posts: 1032
Joined: Sat May 13, 2006 2:40 am
Reputation: 0
Location: Monte Carlo, Monaco
Real name: Michal Kreslik
Gender: Male
Contact:

Postby michal.kreslik » Fri Jan 04, 2008 7:33 am

obx wrote:Michal,

What is the longest bar time frame / tick count you use for FPI pairs?

Thanks,

obx


I'm not using any data series compression at all. My automated FPI framework is fully based on single ticks (new bids and new asks).

Michal

User avatar
michal.kreslik
rank: 1000+ posts
rank: 1000+ posts
Posts: 1032
Joined: Sat May 13, 2006 2:40 am
Reputation: 0
Location: Monte Carlo, Monaco
Real name: Michal Kreslik
Gender: Male
Contact:

Postby michal.kreslik » Fri Jan 04, 2008 7:44 am

cc wrote:Michal,
How do you find the MB/EFX API? The Navigator seems quite unstable.
I wrote a SDK API program to monitor a single FPI ring but the navigator instance seems to crash after about a day, hardly the stuff to which you would want to entrust other peoples money.
Cheers
Craig


Check your internet connection. I've been collecting the data from EFX using a server in a dedicated server room with multiple connection backups from the beginning of Decemeber 2007 and I haven't had a single problem with EFX (and EFX order server) nor NeoTicker stability since then. In fact, I haven't even touched it and it works flawlessly all the time. I'm just regularly downloading the collected data.

When I had the data collecting server hooked up on to my home internet connection, I used to notice there were frequent and strange problems coming up like Navigator database errors, login rejections etc. I was not able to determine what was the cause then. Now it seems the unstable internet connection was to blame.

Michal

cc
rank: <50 posts
rank: <50 posts
Posts: 2
Joined: Wed Dec 26, 2007 7:10 pm
Reputation: 0
Gender: None specified

Postby cc » Sun Jan 06, 2008 8:25 pm

Navigator should be able to cope with disconnections without crashing, though obviously the reliability of the connection is also important.
From what stats I could gather, it seems the EFX data is reasonably efficient.

User avatar
michal.kreslik
rank: 1000+ posts
rank: 1000+ posts
Posts: 1032
Joined: Sat May 13, 2006 2:40 am
Reputation: 0
Location: Monte Carlo, Monaco
Real name: Michal Kreslik
Gender: Male
Contact:

Postby michal.kreslik » Mon Jan 07, 2008 5:27 pm

cc wrote:Navigator should be able to cope with disconnections without crashing, though obviously the reliability of the connection is also important.
From what stats I could gather, it seems the EFX data is reasonably efficient.


I agree, it should be able to cope with disconnections. What is causing the havoc is somewhere in the NeoTicker <> EFX order server <> EFX Navigator trio.

For EFX quote stream quality, check out this post:

http://kreslik.com/forums/viewtopic.php?t=940

Michal

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


Return to “NeoTicker indicators”