Discussion of article "Comparative analysis of 10 flat trading strategies"
Very interesting article (as always with Alexander).
I would like to add a few points (based on developments in the framework of the impulse equilibrium theory):
1. If we talk about flat as a sidewall, we should also talk about the amplitude of price movement (of course, relative to the active movement),
because a sidewall can have a large amplitude and then it can hardly be considered a flat.
2. Structures in a sideways trend differ from the trend by their shape - it is due to these shapes that clear resistance and support levels are formed.
3. The results for all strategies were comparable for a reason:
- traditional indicators are used everywhere,
- and traditional indicators (as well as traditional techanalysis in general) have a global disadvantage:
The frequency of price movements of any financial instrument is constantly changing, and such indicators have no mechanism for identifying the constantly changing frequency.
To quote the previous speaker: "the frequency of price movement of any financial instrument is constantly changing, and such indicators do not have a mechanism for identification of constantly changing frequency". I understand what he wanted to say. And even emotionally agree. But there are big BUTs.
1. Let's look in the reference book. Frequency is a characteristic of a periodic process. The key word is PERIODIC (repeatable). Where do you see a periodic (repeatable) process? I don't see it on the price chart. And I don't see it in the results of the work of Expert Advisors, whether by one indicator or ten at once.
2. If there was a periodic process, and therefore there was a frequency, we could use an amplitude-frequency demodulator.
But still, the problem would be solved in TWO steps, not in one, as in the author of this wonderful article.
3 And since there is no periodicity (frequency), the problem is also solved in TWO steps, but using amplitude-phase demodulation (amplitude is there, phase is there).
Phase is the deviation from the equilibrium position (from the MA). Here's something like this.
To quote the previous speaker: "the frequency of price movement of any financial instrument is constantly changing, and such indicators do not have a mechanism for identification of constantly changing frequency". I understand what he wanted to say. And even emotionally agree. But there are big BUTs.
1. Let's look in the reference book. Frequency is a characteristic of a periodic process. The key word is PERIODIC (repeatable). Where do you see a periodic (repeatable) process? I don't see it on the price chart. And I don't see it in the results of the work of Expert Advisors, whether by one indicator or ten at once.
2. If there was a periodic process, and therefore there was a frequency, we could use an amplitude-frequency demodulator.
But still, the problem would be solved in TWO steps, not in one, as in the author of this wonderful article.
3. And since there is no periodicity (frequency), the problem is also solved in TWO steps, but with the help of amplitude-phase demodulation (amplitude is there, phase is there).
Phase is the deviation from the equilibrium position (from the MA). That's how it is.
Dear Vasily!
The process of financial instruments price movement is generally non-stationary. Within the framework of the impulse equilibrium theory we have identified a structure that is constantly repeating (not just periodically, but constantly and continuously - only the parameters of this structure change). That is why it is possible to determine the period value (or frequency as an inverse value). But within the framework of traditional types of analyses it is absolutely impossible (precisely because there is no mechanism for identifying such a structure).
As for your statement that "phase is a deviation from the equilibrium position" and your suggestion to use a demodulator (and even an amplitude-phase demodulator) - these are rather strange statements, especially when applied to charts of financial instruments prices (i.e. to non-stationary processes). And "in two steps", as you say, the analysis of such a complex (non-stationary) process cannot be solved. That's how it is.
Thank you for your expert opinion. My humble response. The author of the article only worked with amplitude. The first part of the demodulation he did. (Action #1). Good for him. Although these results could have been obtained by simpler methods. By the way, where are the negative results? I don't believe they're missing! Can you email them to me? Now it's on to the second part of the demodulation, the phase, action number 2. I wonder if he'll get it or not. I'm not interested in prompting, if only a little bit. I'd like the author to come to it himself.
Thank you for your expert opinion. My humble response. The author of the article only worked with amplitude. The first part of the demodulation he did. (Action #1). Good for him. Although these results could have been obtained by simpler methods. By the way, where are the negative results? I don't believe they're missing! Can you send them to me by post? Now it's on to the second part of the demodulation, the phase, action number 2. I wonder if he'll get it or not. I'm not interested in prompting, if only a little bit. I would like the author to come to it himself.
Those test results that are in the article are the results of optimisation. Yes, of course, some strategies were not impressive under some parameter settings and conditions. But to go into a deep analysis of the individual pros and cons of each I considered unnecessary within the framework of the article. I tried to show that usually all flat strategies have common strengths and weaknesses.
Very interesting article (as always with Alexander, though).
3. The results for all strategies were comparable for a very good reason:
- traditional indicators are used everywhere,
- and traditional indicators (as well as traditional thechanalysis in general) have a global disadvantage:
The frequency of price movement of any financial instrument is constantly changing, and such indicators have no mechanism for identifying the constantly changing frequency.
Thank you for your feedback. Yes, you are right, I did not accidentally use more recognisable or traditional indicators (although some of them, to filter or identify the signal were trend indicators and were used in the opposite of their application). Before this article was about trend strategies and there I used very complex tandems or interactions of indicators with many parameters in them. Here it was decided to make everything more understandable with recognisable indicators.
Quick question. You said: traditional types of analyses. What are those? TA, FA. And in your opinion, is probabilistic analysis among the traditional ones? Just asking to keep the conversation going.
Of course, both TA and FA, and the probabilistic approach (in the available variations of such analyses) are traditional.
Thanks for the feedback. Yes, you are right, I did not accidentally use more recognisable or traditional indicators (although some of them were trending and used in the opposite way to filter or identify the signal). Before this article was about trend strategies and there I used very complex tandems or interactions of indicators with many parameters in them. Here I decided to make everything more understandable with recognisable indicators.
In my opinion, your approaches are absolutely correct. Thanks again for your interesting work.
Honestly, the article is rubbish. I didn't learn anything new. I'm sorry.
So you're no longer a beginner, you know everything.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
New article Comparative analysis of 10 flat trading strategies has been published:
The article explores the advantages and disadvantages of trading in flat periods. The ten strategies created and tested within this article are based on the tracking of price movements inside a channel. Each strategy is provided with a filtering mechanism, which is aimed at avoiding false market entry signals.
Strategy #1. The Envelopes indicator with the MFI-based filter
The channel borders are determined based on the Envelopes indicator. The MFI indicator is additionally used for filtering of signals.
The below figure shows market entry conditions according to Strategy #1.
Fig. 3. Entry conditions for the flat trading Strategy #1
Author: Alexander Fedosov