Postby Stretchdss » Fri Jun 13, 2008 7:45 pm
I hate to disagree, but I think this can work for a retail trader...
Let me start by saying, that yes, I am new to this forum.
I have read all 67 pages on this topic, and I think I understand what we are talking about...
I believe that stategy can work for a retail trader.
I would like to suggest a few minor "improvements", that may make it possible for a retail trader to use / and profit from this strategy.
1. Let's forget about Interest Swap, that horse is DEAD !!
2. I recall a post a number of pages back that suggested using statistics to find the amount of time that a given ring deviated from "1" by a chosen number of points.
( ie, 90% of the time between 0.9993 and 1.0007, etc.).
I think this is a critical factor in finding rings that CAN BE profitable for the retail "investor".
3. The investor MUST know the Combined Spread of the ring he is trading.
You must calculate the Combined Spread immediately prior to entering a "trade".
If the spread is too wide you should not take the "trade".
If the spread calulation can be automated, it can be used as a filter to keep you out of bad trades. It can alao allow you to get into trades that might otherwise have scared you away.
For example, immediately after News Events...
Once the spread returns to normal, there may be an opportunity to get into trades while the ring deviation is still pretty wide.
This "Spread Filter" might allow you to trade while others sit on the bench waiting for "normal" market conditions.
If, as in the example I gave, the ring spends most of it's time between
0.9993 and 1.0007 (a 14 point range), the "investor" MUST be trading a ring with a Combined Spread of less than 14 (much less would be better), or he/she would have no hope of profitting.
AND, that "investor" should only place orders when the ring is at (or beyond) one of the extremes, and close that ring when it reaches the other extreme.
You MUST know the potential of the ring BEFORE you "trade" it !!!
For example, if the ring has a Combined Spread of 7 (actaully, you can find them as low as 3...), plus a commission of 1 (times 3), for a total of 11. The ring I have illustrated above would return a PROFIT of 3 each time it reached 0.9993 or 1.0007...
please correct me if I'm wrong.
4. I would concetrate my processing power on a FEW rings (maybe 10, with the lowest normal spread), instead of all possible combinations.
If the markets are active, your processor will be working as hard as it can to keep up with all the incoming tick data...
Can this work ? Sure it can !!!
Maybe it will only hit a few "trades" per day...
3 PIP's - nothing to get excited about right ?
I disagree !!!
You give me 3 PIP's, a couple times a day,
with virtually NO RISK ....
I'll quit my day job !!!
I have created a sheet that shows spreads (during London/New York) on the MBTrading platform (I can't remeber where I got it... might have been here !! ).
I used that to find Combined Spreads for a number of possible rings.
Some have remarkably LOW Combined Spreads.
Have a look, it's attached
I need a programmer !!!
I only learned Basic and Fortran back in the 80's
I'm too old, and don't have the time to learn now...
I would like to use the MBTrading platform, so I guess we'll need to program in API...
First, we'll need to use find the extemes as I mentioned above,
then filter by the spread, then make trades when all variable agree.
I am not asking for a hand-out...
I will compensate whoever can help me.
I hope my suggestions are helpful,
and I hope that somehere would be willing to help me...
Thanks to all,
Stretchdss
-
Attachments
-
- FPI Spread Study.xls
- (21.5 KiB) Downloaded 443 times