a trading strategy based on Elliott Wave Theory - page 8
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
Everything is easy on the trend. Both waves and levels. But it is not clear, where the rate will go from the flat. This movement is usually unprofitable for the user.
This link is a closer look at Murray's theory. Of course from the mathematical point of view everything looks very correct and even beautiful (it would even make a good dissertation!:o) The examples given in the files dedicated to Murray's theory, except for postulates about dividing into so many parts and considering these parts responsible for so-and-so things, I have not found anything :o(. Of course it is mentioned, that if the price is higher or lower than this line, the probability of price increase or decrease is about 1%. Besides, there are a lot of exceptions to the rules in this methodology! How in this case can we automate this theory if fractal decision-making exists? Until the theory is calculated in the tester and tester results are published, such as Profitability, Expectation of Growth, Drawdown, and Number of Deals Performed during the Testing Period, statements that the market spends most of the time between so-and-so level do not provide any meaningful information for people who develop MTS! Noise also has its own distribution, in which one can also calculate what is the probability of being within such-and-such a distance of the mathematical expectation. And do we have to make a new theory out of that?
Judging by those conventionally speaking "linear" theories that I am acquainted with, there is in fact an attempt to find what each author would like to find in noise (and this process will apparently continue forever :o)). For example, can someone clearly tell me what is the PRINCIPAL DIFFERENCE between Murray's theory and, for example, Pivot Points, apart from different treatment and calculation of level lines? I don't think so! If we discard all the hullabaloo of each theory in the form of beautiful mathematics, then we come to the most elementary model! Ideally, this model is as follows. There is some process in which there is an oscillation around some conventionally speaking midline. And ALL "linear" theories, without exception, are based on the banal principle of continuation of motion. That is, the formula known since the second grade of school S=V*t (S-way, V-velocity, t-time), is the basis of all existing "linear" theories and "linear" theories to be invented in the future! So what does PivotPoints and Murray do? It takes information about price movement in the previous time interval and calculates points by linear formulas (by linear I mean the fact that the formulas use only the simplest operations such as +, -, *, without using any trigonometric functions and exponentiation). Then you don't need to fill your head with beautiful theories, but just calculate these same coefficients for placing orders using the formula of second grade school (I do it in my strategy using the formula that makes absolutely the same sense as the one mentioned above). That is, using linear formulas you can for example predict where the price will be in the next same period of time, while it will keep the movement in the same direction, or change it to the opposite! This algorithm, which may be very simple, is used in the tester for calculation of conversion factors. And I can tell for sure that working efficiency according to any simplest formula for calculating points will be quite as good as working efficiency according to some Pivots Points or Murray's theory, since they all contain the same trivial principle of predicting movement based on previous information about movement, the only difference being that the beautiful theory will be wrapped in a beautiful wrapper! And in general until people finally understand that FOREX is noise with its consequences, the most ingenious theories will be born, which "describe the market very accurately" :o))). And people will study them and even pay money for training with some fee-based techniques. In general, it's like in religion. If you believe - it helps, and if you don't - it won't. I will listen to your objections with interest.
For questions, perhaps not in the order they came up.
1. The most important thing to accept is that any result only makes sense within the framework of the problem set. The task itself is set on the basis of a certain group of axioms and postulates. Those who have been educated in research methods and validity assessment know that the objective setting represents at least 90 per cent (80 per cent) of success or failure. And only practical evaluation of results can show how true are the theoretical assumptions and how much the result is natural (with what probability the system built with similar approach and based on random (pseudo-random) noise will give the same result.
2. Regarding Murray levels - they just fit as one of the solution methods for finding constraints of the problem I was setting - nothing more. I think these methods themselves were "pulled" from some wider problem statement and are not much usable on their own, although in combination with other analysis methods they are rather useful. In principle, you can use Pivots as well (just the statistics will have to be recalculated ;) ).
So I use Murray levels only as one of limitations for projections. The experience shows that some of the levels, known before the price reached them, are a very reliable estimate for support/resistance levels. And it has been rightly observed that deciding to open a position near these levels will bring the trader minimal risk - isn't that useful? The question remains as to the timing and which of the levels at any given time. These estimates are obtained by projecting movement by linear regression method - again we may take another one - the main thing here is the principle and again we will have to recalculate statistical estimates for application of another method.)
3. The question about the direction of a flat breakthrough - for me it is not a question at all because the formulation of the problem does not suppose any flat (here I am deeply grateful to Tactics Adverza for the idea - "flat is a part of a higher order trend" (it is their original interpretation) - in this formulation there are simply moments in time when forecasting is impossible, in this case we simply hold the previous position, if it exists. Entry into the market is made at the moment when it is estimated that non-random prediction is possible by a trend (or more often by a counter-trend - it depends on the estimation) and further the probability of a trend movement is estimated, its goals, until the goals are reached the trend is considered as continuing and a trend position is held, increasing the probability of movement and increasing the market participation. That is, within this task there is only the notion of "trend" and it is not defined in the usual way, but only quantitatively, i.e. a trend is an excess of movement probability to one side, as a consequence - the possibility of non-accidental forecasting.)
3. As for the rest: of course one can accept the efficient market theory and believe that the whole market is white noise and a successful prediction is an accidental guess. It's not the worst approach, but then you lose the opportunity to set a problem that involves a solution in the sense of non-random prediction and what kind of prediction can we talk about? Only guessing. Of course you can do it that way - then capital management rules come first: otherwise you cannot get a positive result - and note that basically all money management rules are built on the premise that trades are independent and 50x50 probability. Although the approach I trade assumes that all trades making up one trend are dependent - hence slightly different statistical estimates ;).
4. Regarding the basics of constructing both "linear" and "non-linear" All their mathematics is based not on the formula of continuation of movement S = v*t, but on the Taylor series (as well as almost all applied mathematics and harmonic analysis, too - it is simply a representation in the analytical, easily integrated by hand functions and the same Fourier series is a special case of the Taylor series), while this formula itself is a consequence - that is a linear approximation. If you take into account terms of higher order, you will get more exact approximation (the second term of the series will be acceleration, just in case I remind you more exact formula
S = v*t + (a*t^2)/2 ;) ).
But there is another term in Taylor series, which allows to estimate method's error - don't forget about it too.
All the above is IMHO, of course, but the possibility of quantitative estimates makes such an approach preferable for me and, after all, those who wish may make quantitative estimates themselves.
Good luck and good luck with the trends.
Read the previous post carefully - it's about interpreting the results within the task at hand ;).
And who do you think should do it all? Who else but you, the MTS developer (as far as I understand) needs it. Or someone ordered you MTS on these principles - then ask him ;). Basically, the algorithm is there if you are interested - do it, if not, then don't.
See the very first post about Hirst's criterion, and more - why ?
Good luck and good trends.
Apparently we use profoundly different trading techniques. You "Tactics Adverza for the idea -" flat is part of a higher order trend", and I'm just going head-to-head without any extraneous ideas. (I play, so to speak, in shallow water, i.e. in the current flat, to which the noise model fits quite well).
So as far as I understand you're using Murray as part of a larger strategy that you're trading manually? Yes or no?
I don't do the work of tracking breakouts as it is a very thankless task. I deal only with calm segments that suppose a return of price from points of maximal deviation back to the central part and further to the opposite side.
By the way, if it's not a secret, how many percent of a deposit in a month are growing by your strategy?
My maximal drawdown during 1.5 years without exceeding 20% is averaging 18% profit a month. What is your average profit?
Yes, although not entirely manually - there is no need to constantly monitor the market and therefore experts are not needed.
I don't do the work of tracking breakouts as it's a very thankless task. I deal only with calm segments that suppose a return of price from points of maximal deviation back to the central part and further to the opposite side.
It is very similar to the approach I use, but not only on quiet sections. I do not use Adverse Tactics as such - I just like the idea of determining the trend.
By the way, if it's not a secret, how many percent of a deposit in a month are growing by your strategy?
My maximal drawdown during 1.5 years without exceeding 20% is averaging 18% profit a month. What is your average profit?
In general, it is not a secret - it depends on the level of risk that a trader is ready to take. For example, from January till the end of February I have almost doubled my deposit on real account (more exactly 93%), trading with minimal risks, and for monitoring I used a demo account with maximal risks - there I traded almost 450%, but I would not risk so much on real account :) - These were the best figures, while the average was of the order of 40%.
Good luck and good trends.
I have already written about it above. I use the Standard Deviation indicator from MT4 standard delivery set. I know there are many different indicators, one of the most popular is Juice (I used it too, but later I refused it because of its redundancy). But anyway all indicators are based on analysis of current market deviation. Something technically better is unlikely to be invented.
I have already written about it above. I use Standard Deviation indicator from MT4 standard delivery set. I know there are a lot of different indicators-filters of the flux, one of the most popular is Juice (I used to use it too, but then I gave it up because of its redundancy). But anyway all indicators are based on analysis of current market deviation. Something better technically is hardly possible.
In general, we should remember that the standard deviation is calculated with respect to a predicted value. The algorithm in the standard delivery counts relative to the muving. So you voluntarily or involuntarily take the approximation of the movement by a muvin of a certain order.
Good luck and happy trends.