Utility trawl - utopia or reality? - page 3

 
TheXpert писал(а) >>

Um, there is the concept of Trailing Stop -- a pull-up stop and Trailing Profit. What is a trailing stop on "entry signals"? Can you be more specific?

Well, let's assume that the entry signal for you is a MACD crossing with zero from the bottom to the top. Let's set the Stop price at, say, 30 points. Then suppose that the MACD starts increasing and the stop moves behind the price with a certain speed, I think it would be better to increase the Stop. Let's say that after a while the MACD went down. The stop moves closer to the price. I.e., when trawling, we look not at the price movement but at the behaviour of our signals. I will draw a rough picture now. But this is for SELL.

 
infinum13 >> :

Well, let's assume that the entry signal for you is a MACD cross with zero from the bottom to the top. You set a Stop, let's say 30 pips. Then let us assume that the MACD starts increasing and the stop moves behind the price with a certain speed, I think it would be better to increase the Stop. Let's say that after a while the MACD went down. The stop moves closer to the price. I.e., when trawling, we look not at the price movement but at the behaviour of our signals. I will draw a rough picture now. But it's for SELL.

Have you tried to stop trawl on MACD reversal? If trending - the next top should be lower than the previous one.

 
TheXpert 07.04.2009 12:29

They can, in theory, but do they give them away? That's the question, in principle.

Theoretically they do. But in general, by conviction, it seems to me that the case of trawls being able to replace an exit point is from the same line as Graali. I mean, it didn't work out for me. They should be, but only as a reserve parachute - the main thing is not to crash, not a soft landing.

 
locol91 писал(а) >>

Have you tried stopping the trawl when the MACD reverses? If there is a trend, the next top should be lower than the previous one.

I do not trawl orders at all. This is my speculation and nothing else. MACD is the simplest example.

 
TheXpert >> :

Is there such a trawl that increases the expectation of trades for a short-term strategy?

This formulation is probably not quite correct, if we ask simply - is there a universal and useful trawl?


Haven't seen any so far, but maybe I missed something?

It exists, but it can hardly be formalized for MTS. The market is constantly changing and we hardly can "mechanically" detect these changes in time to make a decision at the right moment to keep trawling or close it.

I have checked a lot of methods of trawling through history, and none of them has the required efficiency that I need to achieve in any market. The worst disease is premature stop-loss due to a bunch of different reasons. Most often this happens because pullbacks are rarely of any one size. In addition, there is no dependence on the size of pullbacks at the beginning of the trend and at the end.

If we teach MTS to predict the size of the pullback, then we could make a "variable trawl", but it's complicated and it's only one of the "dots" solution.

The most reliable method I know - to keep the stop at the level of the extremum "n" (for NT) until the breakdown of the minimum preceding the extremum "n+1". Long trends are obtained using this method amazingly, but as soon as I get into a "mess" this method fails. All other trawl methods I do not even really understand how they can be useful.

 

Option:

A distant trawl stop, moved at a "respectful" distance and intended solely as a safety net, I would say a no-brake option. The basis of such trawling - entry only at strong signals suggesting a good trend and closing not at the trawl stop, but at the strong reversal signal or at a strong target level.

It turns out that this variant does not really relate to trailing in the sense that it is usually understood. But IMHO it is at least not worse.

 

If the stop is far away, what's the point of trawling it at all? What difference does it make, a little more, a little less...

It makes sense to trawl something that actually works according to a given strategy, not an abstract insurance policy.

 

I think the MO might be increased by changing the stop and take to the optimum trawl

something like this)

i.e. not just simple stops and takes but an optimal system of stops and takes, and the trail separately)

 
Jingo >> :

I mean not just stops and toes, but an optimal system of stops and toes, and the trawl separately)

It's a blunt fit.

 
I think this is a good trawl: sl=x p. When the price gets further from the opening price by x p., then we put sl to breakeven (at the opening price), next level of sl is pulled up by x p. when the price is already at 2*x p. I.e. we only pull up the sl level when there is 2*x p. between the price and the sl level, to the x p. level.
Reason: