MQL4 - automated forex trading   /  

Forum

Read Financial Magazines in MetaTrader 4 and MetaTrader 5!

WHIPSAW, TREND, or COMBINED?

Back to topics list  | 1 2 To post a new topic, please log in or register
avatar
75
EuroTrader 2010.05.01 08:48
 
By using Differential Calculus I can easily locate the bars for the LOCAL MINIMAs and LOCAL MAXIMAs on a 30 min EURUSD chart. They are basically the LOCAL TURNING POINTS. The only problem is how far the turns will go. It could WHIPSAW or just keep on going as a long TREND after the turns.

So I've developed one WHIPSAW EA and one TREND EA. The signals are totally opposite depending on whether the next few bars are a WHIPSAW or a TREND. For WHIPSAW the EA sells at next bar after the LOCAL MINIMA but for TREND the EA buys at the LOCAL MINIMA, and the signals are reversed for the LOCAL MAXIMAs.

Now one can easily imagine my dilemma of which EA to use. If I used WHIPSAW EA I made money if the market had no trends but lost massive when the market suddenly went on a long trend. When I used TREND EA I made good money if the market trended but lost all when the market suddenly whipsawed for a long time.

(A long TREND is equal to a few WHIPSAWs and they eventually cancel out each other as if the bloody market has no gain or loss money-wise.)

So I've combined the two EAs. I've been running a few days the COMBINED EA and my account balance is basically stuck at almost where I started. The bloody EA is behaving as if a perfect hedge.

Any thoughts on this topic and please help me out of this doldrums? I can give away my simple and robust EAs free to anyone who can give me a profitable idea out of the HEDGE I am in now!






 
Testing Expert Advisors on Non-Standard Time Frames

Testing Expert Advisors on Non-Standard Time Frames

It's not just simple; it's super simple. Testing Expert Advisors on non-standard time frames is possible! All we need to do is to replace standard time frame data with non-standard time frame data. Furthermore, we can even test Expert Advisors that use data from several non-standard time frames.

avatar
712
1005phillip 2010.05.01 16:39 #
 

The only problem is how far the turns will go. It could WHIPSAW or just keep on going as a long TREND after the turns.


Your dilemma is fundamental of all financial instruments that have non-zero second derivatives.

Namely the question "is this current change in price direction the harbinger of a reversal or merely a retracement?"

Answer that question with a concoction of mathematically expressible analytical equations and you'll be the master of every investment market in existence, not just forex.
The bloody EA is behaving as if a perfect hedge

Since a perfect hedge is basically tantamount to not being in the market at all (aside from the equity loss thru spread plus that of the 2*rollover fee accruals) this should result not be surprising. Your strategy is basically saying it has a 50/50 chance of correctly predicting reversal vs. retracement and when the odds are 50/50 you either sit out (no trading) or dive in with a hedge (and lose money to broker).

Hedge do not generate profits. If you have a strategy that generates profits from hedging then your strategy will be all the more profitable (by an amount equal to that of the rollover rate on the instrument) by just closing and reversing. Hedge merely keeps you "in" the market, not creating profits. It is great for account managers and brokers, i.e. anyone who profits by your lot-turnover rates thru spreads and commissions, but thats about it.
 
avatar
Moderator
2029
gordon 2010.05.01 17:49 #
 
EuroTrader:
...idea out of the HEDGE I am in now!

My own opinion about hedging (and my most valuable contribution to this forum to date) -> http://forum.mql4.com/22859/page5#269081 :)


 
avatar
712
1005phillip 2010.05.02 05:00 #
 
gordon wrote >>

My own opinion about hedging (and my most valuable contribution to this forum to date) -> http://forum.mql4.com/22859/page5#269081 :)


You have a true gift gordon in being capable of enduring the kind of absurd abuse directed towards you in that thread.

I too have hedged, lord knows I spent a year solely dedicated to optimizing strategies around it, then I stepped back and realized in the ever-bigger picture that while it can work it is entirely needless in comparison to standard entry/exit/entry routines.

The only time it makes sense to hedge a currency pair is when you are trading a cross, say the AUDCAD, where your equity at risk (and resultant stoploss value) depends on both the price of the principal financial instrument (AUDCAD in this case) as well as the price of the counter currency pair (USDCAD in this case). In this case it makes sense to open a hedge in the counter currency pair (mindful of the leverage differential of course) such that your stoploss price retains its true equity at risk valuation.

Not sure how common it is to come across folks here who have passed their series 3 exam but in preparing to become a registered CPA with the NFA I passed my series 3 and that was when/where I came to learn about the purpose of hedging as well as the realization that my efforts to use it had truly been in vain. I could have accomplished the same effect but made more profits in the meantime on the same trades. Just gave away swap to the broker, nothing more was gained.

 
avatar
Moderator
1937
BarrowBoy 2010.05.02 22:13 #
 
EuroTrader:
So I've developed one WHIPSAW EA and one TREND EA... Now one can easily imagine my dilemma of which EA to use.





ET

See this EA for thoughts on trend/range detection http://codebase.mql4.com/5292

FWIW

-BB-

 
avatar
Moderator
1937
BarrowBoy 2010.05.02 22:17 #
 
1005phillip:


The only time it makes sense to hedge a currency pair is when you are trading a cross, say the AUDCAD, where your equity at risk (and resultant stoploss value) depends on both the price of the principal financial instrument (AUDCAD in this case) as well as the price of the counter currency pair (USDCAD in this case). In this case it makes sense to open a hedge in the counter currency pair (mindful of the leverage differential of course) such that your stoploss price retains its true equity at risk valuation.

AUDCAD/USDCAD isnt a great choice for even incidental hedging use - the correlation pattern isnt good enough :(

At least you haven't confused covering with hedging :)

Good Luck

-BB-

 
avatar
712
1005phillip 2010.05.02 22:38 #
 
BarrowBoy wrote >>

AUDCAD/USDCAD isnt a great choice for even incidental hedging use - the correlation pattern isnt good enough :(

At least you haven't confused covering with hedging :)

Good Luck

-BB-



BarrowBoy I think you misunderstood my post.

If your account currency is USD and you buy 1 lot on AUDCAD with 100:1 leverage then you are also going long 0.01 lots of USDCAD at the same time. (if you aren't convinced or don't see this to be self-evident then I recommend working thru the calculation of what defines tickvalue for cross-currency pairs)

If you only want to go long on an AUDCAD position and not have your position value affected by the fluctuating USDCAD price then you need to short 0.01 lots of USDCAD at the same time you opened your long position.

This is true for every cross-pair (although whether you need to short or long the counter currency pair changes). However in reality, with the leverages involved, it is not an option to open the hedge position with the appropriate lotsize.

Hedging is intended to lock-in prices, that is where the risk reduction comes in. If you are a farmer who wants a guaranteed corn price this fall at harvest you sell futures on corn in the spring and buy crop insurance. If corn prices rise then your futures decline in value but your crop rises in value correspondingly, etc.

 
avatar
75
EuroTrader 2010.05.03 20:51 #
 
One way out of the hedge pickle is a well-researched placement of STOP LOSS and TAKE PROFIT. Lets say the upturn has started and a LONG position is taken by the TREND EA. Then immediately or after a bar an equal but SHORT position is taken by the WHIPSAW EA. So I am now in the HEDGE. There are three possibilities.

1) If the UP TREND reverse I can close the hedge with either a small profit or small loss.

2) If the UP TREND continues the STOP LOSS for the SHORT is triggered and the SHORT is closed. If the UP TREND keeps on continue the TAKE PROFIT for the LONG is triggered and the LONG is closed and I have a serious profit.

3) I have a potentially big loss if the UP TREND reverses after taking out the SHORT as my LONG is now out of the HEDGE.

To get the levels of SL and TP right is the crucial factor to turn my EA into a profitable EA.


 
avatar
712
1005phillip 2010.05.03 21:17 #
 

EuroTrader that strategy will cost you less if you replace the hedge approach with a Buy Stop/Sell Stop straddle strategy.



Being "in" the market with a hedge just means you paid 2x the spread for the priviledge of then paying 2x the roll-over rates until such time that you finally exit the hedge.

A straddle costs you none of that because pending orders don't cost money until they are triggered, and if you split your hedge into pending orders then you aren't "in the market" until the market has decided whether it is merely retracing or actually reversing.

In your case you would use your the indicators which are saying "buy now!" to actually just set a pending Buy Stop at the price level you were going to set your stops for the short position anyways. Likewise when your indicators are screaming "open a short now!" you would instead open a Sell Stop order with the price placed where you would have set your long position's stop anyways.

 
avatar
75
EuroTrader 2010.05.04 04:31 #
 
Phillip,

Thanks for your considerable thoughts. I have traded that straddle almost a year back in the 00s and one thing I do not like is if the market is slow I was out of the position most of the time.(Also one thing about the straddle is there are too many outcomes.)

My aim now is I want to be in the market all the time holding a position. Analogy is like in the water all the time riding the waves. I want to be bobbed along all the time, not drowning or idly lying on the beach. My current EA, like a life jacket, now let me do that and, thank God, for the start I am not losing money. Any way I used 25 pips SL and 70 pips TP and I made nearly 200 pips last night in 30 min EURUSD. Once in a while the market fits in perfectly with my system.

In the chart the blue arrows are WHIPSAW signals generated by the WHIPSAW EA and the magenta arrows are the TREND signals by the TREND EA.There were two big down-moves and my automated system achieved the most profitable outcome, the OUTCOME 2 mentioned in my previous post. That two moves alone gave me 140 pips, two TP points.

The WHIPSAW wins and the TREND losses canceled out each other while the WHIPSAW losses are just too small but the TREND wins are big enough to make the EA a profitable one for that instance.












 
avatar
Moderator
2029
gordon 2010.05.04 04:43 #
 
Bottom line is that if u always open the 'net position' (with no hedging), you would make the same profit, but cut down on spread & swap.
 
Back to topics list   | 1 2  

To add comments, please log in or register