Machine learning in trading: theory, models, practice and algo-trading - page 537
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
good rule). It automatically follows that the price takes into account not only the present but also the foreseeable future. From this it automatically follows that any price prediction is absolutely impossible.
That is why I always write that only inefficiencies, which periodically appear here and there, then disappear just as suddenly, work :)
The theory of an efficient market... it doesn't say anything about the future, but the definition says that if events have an instant effect on the market and it's efficient, then none of the participants has an advantage, yes (with strong efficiency there's also medium and weak one).
That's why I always write that only inefficiencies work, which periodically appear here and there, then also suddenly disappear :)
The theory of an efficient market... it doesn't say anything about the future, but the definition says that if events have an instant effect on the market and it's efficient, then none of the participants have an advantage, yes (if the efficiency is high, then there is a medium and low one).
It takes into account EVERYTHING - that means EVERYTHING. And if something is not taken into account, such as the future, then it means NOT ALL, which contradicts - takes into account ALL. That is, the future, by its very definition, is unpredictable in principle, since everything that is known has already been taken into account, and only random components remain unaccounted for. If you think that you can find something like that in the market, then, again, this means that you think that the market hasn't already taken EVERYTHING into account. Then you are lying.)
And you're talking about an efficient market.
Taking into account ALL means ALL. And if something does not take into account, for example, the future, it means NOT ALL, which contradicts it - takes into account ALL. In other words, the future, by its very definition, is unpredictable in principle, since everything that is known has already been taken into account, and only random components remain unaccounted for. If you think that you can find something like that in the market, then, again, this means that you think that the market hasn't already taken EVERYTHING into account. Then you are lying.)
And you yourself are talking about an efficient market.
Well this is a hypothesis... and most likely it's wrong in the general case, because there are no absolutely rational participants in the market who instantly react to a market event
that's why it's usually like this :) either red or green
But very importantly, the hypothesis assumes that on average it is impossible to outplay the market, i.e. on average all participants will be in zero, which is reasonable. But taking into account that income distribution is usually uneven, we have a chance, it's just 5-10% who earn, but even more chance to earn nothing or have large losses :)
I`ve got some information, I`m too lazy to write. I have been doing this for many years and I'm still working on it.Well this is a hypothesis... and most likely it is wrong in the general case, because in the market there are no absolutely rational participants instantly reacting to the market event
that's why it usually goes something like this :) either red or green
But very importantly, the hypothesis assumes that on average it is impossible to outplay the market, i.e. on average all participants will be in zero, which is reasonable. But if we take into account that income distribution is usually highly unequal, we have a chance, it's just 5-10% who earn, but an even bigger chance to earn nothing or get large losses :)
Here's some info, because I'm too lazy to write. It's actually very useful to understand this.Anyway, I've been interested in efficient markets and have read a lot on the subject. Most modern researchers believe that today's market is very close to efficient. Including, about 5 years ago at the RTS conference there were reports on this subject made by quite famous specialists (unfortunately I forgot the last names - 5 years, however).
Perhaps the only inefficiency is the phase delay in the response to external influences. What you have drawn. However on a step it looks spectacular, but on smoother influences, if there is no prior information, you can catch it only when the train has already left.
In my experiments I managed to achieve a stable result with new data (to guess the direction of a new bar on five-minute eurusd with accuracy of 51%-52%), but the problem is with the spread - the model will be profitable if the spread is no more than 2 points. If you trade by the model only in the evening hours the potential profit is a bit larger, but the spread at this time is also larger.
By intuition, an interesting self-regulating system is formed - if traders trade on the plus side, the spread will increase. If everyone is losing money in general, the spread decreases due to competition.
There are some constant dependencies inside spread, but the dealing dealer regulates spread so that these dependencies are available only to the market makers.
It turns out that opening market deals is an invitation to oneself and it is necessary to implement trading strategies with pending orders, then there may be some chance for a grail.
I'm a fucking programmer. For four hours I tried to make an AD indicator for MT5 using CDs, but I kind of did it. This is a mess, comrades. I got lost in three lines :-(
You could have just searched
It automatically follows that any price prediction is absolutely impossible.
It follows from this conclusion that someone's logic is lame on both legs or even deprived of them))
And you could have just searched
What a village... I did it under MT5.... dark........
What a village... I did it under MT5.... Darkness.....
10 minutes for redesign instead of 4 hours.
10 minutes to redo instead of 4 hours.
That's the kind of programmer I am. No one can explain it, until you get the hang of it... Anyway,people!!!!! Listen up. Here comes the important long post......