a trading strategy based on Elliott Wave Theory - page 271
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The plot is indeed complex, and the rules I mentioned are applicable with some (and not so little) stretch. Although channels often appeared at me at two extrema, i.e. I cannot say beforehand that a short channel cannot exist till the 500th readout. It even seems to me that it should have been, only narrower and practically horizontal (approximately 1.26 - 1.28), but I will not "start the tractor" to check it :).
I did not come across any canonical definition, I interpret it as a new maximum (minimum) at the end of a trend. Corrections can be very different.
But then at least share your achievements in trend detection.
The matter is not that, I warned from the start that some examples will not convince me of anything. Only statistics may convince (or rather a desire to check it myself). For example, when testing each order, I wrote values of the analysed parameters at the moment of opening into a file and then a profit/performance chart was plotted. Then we usually found out at once that we could efficiently separate samples of positive and negative profits only by trashing a large part of orders.
As far as trends are concerned, I do not have much to brag about. Moreover, it is not a hot topic for me now.
I test the system components separately as a matter of principle. I do not "mix" tactics/strategies with calculated parameters, when evaluating the effectiveness of components. I compile statistics on the components separately. As I wrote before, if Hearst should show continuation/change in movement, then I test it by estimating the lifetime of channels (no orders during testing of components). Since there are a lot of complex sections, I have to check everything "by eye". The Hearst statistics shows quite the opposite, i.e. excellent results.
Still I started the "tractor" (something out of the mood :). The channel has had a nice time
And what is the Hearst score for the longest channel? And if this is the first sample (or even the second), what channel has time to draw? A stable channel should show just the viz.
You have to go into the code and pull it out - in this version it doesn't count at all. And the topic has been frozen since last September. That is, something does not pull me to this feat :). Yes, and I do not see the point. A comparison of calculation methodologies? But I will not return to this approach now anyway.
You have to go into the code and pull it out - in this version it doesn't count at all. And the topic has been frozen since last September. That is, something does not pull me to this feat :). And I do not see the point. A comparison of calculation methodologies? But I will not return to this approach now anyway.
But the thing is, if your system shows a long channel as a channel that can be trusted, then it is fundamentally wrong (I "forcibly" drew it and showed that it cannot be trusted, my system gives only one, the most reliable one). This channel is about to collapse. Or do you assess this channel in any other way?
My profits/losses have been bound to the levels of SPE. That is, the result of the deal actually showed the quality of the channel.
I don't think this question was asked before :) . It is difficult for me to answer it because the algorithms are really different and we cannot always find analogues of one concept in another one. And the concept itself should be carefully clarified beforehand.
Yes, with the help of the aforementioned statistical tortures I was just trying to find a means of estimating the quality of the channel.
I'm not sure it's a proper assessment of channel reliability. You can "easily" get a loss in a reliable channel, especially if the channel has a large spread.
I don't think this question existed before :) . It is difficult for me to answer it, because we actually have different algorithms and we can not always find analogues of one concept to another. And the concept itself should be carefully clarified beforehand.
I thought I would have time to reformulate it, only a few minutes passed, but I had no time and there was no sense to start a new post. :о)))
After evaluation of channels, I'm looking ahead, no special algorithms are needed and I'm testing control logic at the same time.
Yes, with the help of the aforementioned statistical tortures I was just trying to find a means of estimating the quality of the channel.
I took a different way and derive an empirical estimate (formula) of channel lifetime based on estimate of parameter statistics. No deals at all. Deal management is the next level of the system, and the lower level is a good model of dynamics.
У меня профиты/лоссы былы привязаны к уровням СКО. То есть результат сделки фактически показывал качество канала.
I'm not sure this is a good estimate of channel reliability. You can "easily" get a loss in a reliable channel as well, especially if the channel has a large spread.
One of the 40 indicators reported inside or outside the channel an order was closed. These cases were marked with different icons (and colours) on the diagram.
And this was ((C) Ecclesiastes :). Another one of the 40 indicators was the lifetime of the channel :).
У меня профиты/лоссы былы привязаны к уровням СКО. То есть результат сделки фактически показывал качество канала.
Не уверен, что это правильная оценка надежности канала. У Вас лосс «легко» может получиться и в надежном канале, особенно, если канал имеет большой размах.
One of the 40 indicators reported inside or outside the channel an order was closed. These cases were indicated on the chart with different icons (and different colours).
And it was ((C) Ecclesiastes:). Another one of the 40 indicators was the lifetime of the channel :).
I think it's easier for you to list what it wasn't. :о))))
This is the minimum lot, no reinvestment. That is, the final surplus is about 3600 pips. And in fact they were made after January 2004, i.e. for 2.5 years (my point is that the history before 2004 looks more and more unclear to me).
I was not satisfied mainly with the insufficient number of deals (from 400 to 600 in various versions). Of course, I could not fully unravel all this heap of indicators and their combinations. Nevertheless, I think it is impossible to reach the conclusion that the score cannot be improved without significant reductions in the number of deals. And this is certainly unacceptable.