Market patterns - page 4

 

This thread is probably a good example of why this is the case:

Алготрейдеры - засранцы: шифруются.

In that context, being an asshole is a lot more correct.

 
223231:
I agree. Forex is influenced by too much information and it is very patchy, so it looks very much like a random move. So far, it seems to me, that it is impossible to determine the direction for the next 10 hours, you can only determine the approximate character of this movement. Although I would love to hear from someone who doesn't think so (and not just IMHO, and the method I would like to see).
Note, I have not said a word about the direction. To know the probability of continuation of a movement in general and to be able to predict the future direction of the price at specific points in time - are different things.
 
C-4:
Note, I have not said a word about the direction. Knowing the probability of continuation of movement as a whole and being able to predict the future price direction at specific points in time are different things.

I would like to point out that I agree with you.

You don't have to use complex models for this (as soon as you start applying them in the market they don't work anymore).

Any mathematical model can be so beautifully designed - it is of no use in a real situation.

It should be simpler than that.

 
C-4:

I'm referring to Hearst's value again (it's just that whenever we talk about the probability of a trend continuation/reversal, we're actually referring to price determinism). That's why I think it's appropriate to mention this value once again (but only in passing) in this thread.

As for the method itself, it's a custom non-parametric numerical method based on the zig-zag indicator. The mathematics itself is specific, I don't think it's worth discussing here. The only important thing is that the calculation results coincide with the test data series with a decent accuracy. So the coefficient calculated for the series with the known value H=0.30 turned out 0.34, for rad 0.50 turned out 0.51 for rad 0.70 turned out 0.70 (i.e. the more the trend, the less the error). However, as far as I know, there are functions in R language, in particular in "pracma" package, which work with no less decent accuracy, but use tens of times less computing resources. Therefore, the method itself is secondary, the only important thing is that the values obtained on the test samples are close to the expected ones, and therefore we can assume that the value obtained for the markets is also close to the true one.

Different markets were tested. In this regard, they show characteristics similar to each other: all trendiness values are in the range of 0.5 - 0.54, but there are some other statistics, which differ very much. In the complex calculation of H, I also obtained some other values, which are very different from market to market, but what they mean I do not know yet. I am working on it now. However, there are first guesses. So I think I am close to a numerical definition of the Noah/Joseph effect, as well as a coefficient showing the noisiness of the market (if it is true, Forex is one of the noisiest markets).

Imha, this is an error in assumptions. I.e. a consequence of memory in volatility, clustering etc. Just otherwise this test, actually a profitable TS on any markets and without filters. Fantastic. Anyway, it would be interesting to discuss this test in more detail, but apparently this is an offtopic here
 
Avals:
imha, this is an error in assumptions. I.e. the consequence of memory in volatility, clustering etc. Just otherwise this test, effectively a profitable TS on any markets and without filters. Fantastic. All in all, it would be interesting to discuss this test in more detail, but apparently this is offtop here
If you mean the ideal TS for all markets, it cannot be described here, it does not exist. More specifically, on certain intervals, you can consider such, with a certain margin of error.
 
pantural:

used to read diagonally.

"You also read the robust strategies thread diagonally, otherwise you would have found the answer to your question.

St.Vitaliy2012.10.12 08:57

Excellent formulation of a simple question - why?

i'm increasingly seeing advertising on simple sites with signals via sms. some are trying to write books on the market and how cool and profitable to trade... but if they have reached such a professional level, is it impossible to use their system to consistently bring profit to themselves, their future clients and their families) i sometimes cannot understand... how come? trading is already so close to the money, that you just take it, learn and earn. And the opportunities and the ceiling are unlimited, you can earn millions of rubles. But no. They start earning money on all sorts of trifling things, like books, trading signals via sms and web money, robots... I understand it for designers, webmasters, creative people... they are poor and destitute even as adults. Their brains are tuned to creativity, muse and passion for the process, not for making money. They have no desire to understand the whole financial world. But this is trading, finance, futures, stocks, options! Why all these books? SMS messaging, pathetic excel robots. What for? To earn 30 kopecks on this nonsense? Why are these smart guys concentrating more on that? They are not trying to develop their trading system, not trying to develop fundamental and technical factors, their psycho-type and psychology? And if the signals are really right, why sell them for three rubles, why write books when you can earn tens of thousands on them?

Please explain, so I do not have to return to this question for myself.

 
iModify:
If you mean an ideal TS for all markets, you will definitely not find a description of it here, there is no such thing. More specifically, on certain intervals, you can consider such, with a certain margin of error.
That is what I mean - there is no such TS, consequently there is no test to distinguish "randomness"/randomness from non-randomness. More precisely, such a test is Equity of Profit system. But there are no eternal and universal systems, and they filter the market.)
 
Avals:
This is what I mean - there is no such TC, therefore there is no test to distinguish "randomness"/CB from non-randomness
Wrong, there is not one but a whole family. You can test for potential non-random sequences over a wide range of relationships between data of the same series, as well as a large population, this is the basis of proffesional approach. In principle, the questioner is asking the right questions, but the paradox is that there are simply no unambiguous answers to the right questions, and even if there were, a publicly stated answer by someone unreasonable would immediately negate its potential, so it is impossible to answer, any public answer would be wrong for that reason.
 
C-4:
Every pattern is based on the principle of either trend continuation or trend reversal. No matter how many patterns there are, they will all use the same effects. To illustrate this, take any two trend robots based on different principles. Run them on the same symbol - their correlation will be high. The reason is that although they both use different patterns, they nevertheless make money thanks to the common notional H. It's that simple.

I understand what you mean. But this is a very gross simplification, as a rule such brutal patterns are "stolen before us" already, or on the assumption that they are so obvious, traps are built, from figures at higher levels of the food chain.

It's like when talking about music, for example, to generalise it down to loud silent and absent. But there is classical, there is pop, there is heavy metal, there is techno and so on. There are also many structures, formations, patterns on the market, each group of which signals with more precision the probability of some further behaviour than the MO groups of all FI groups under the name of "trend". That is why "trending" in general is so unobvious (3%). If we can recognize major micro groups of such formations at corresponding FI (indices, etc.), knowing probabilities for each group of formations or particular elements, we will have a considerable statistical advantage. Well, of course such formations should be found and found automatically. I'm working on it. It is a multi-stage technology that is not very elegant yet, a hybrid of statistical processing, neural networks and human analysis but its essence can already be seen. And it is clear that publicly known formations have already been nibbled and are almost devoid of profit potential.

Sabjmaker is looking in the right direction, but I'd say it's all very difficult technically. Matan, statistics, functional analysis, expert systems, neural networks, etc. etc. It's AI, what can I say, the most complex thing on earth.

 
You do not need to develop your own platform first, there are ready-made ones with a wide range of functions, including access to Level 2 and so on.
I am more interested in advice, how can you analyse Level 2 in forex, it is a fragmented market, each liquidity provider has its own data and everyone is different. Even if we take exchange-traded forex futures, this information is not quite adequate, because it is a derivative of the instrument and the deciding factor is where the underlying instrument will go. I have thought about this problem a lot, but I still do not understand how it may influence it. In the stock market I understand how lev 2 affects the price, but in Forex? With its dispersion? I would like to know your opinion on this matter.