Combining multiple indicators in an EA - 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
Плохая формула. ;)
Объединение индикаторов не означает их пересечение.
U
Put the horns down everywhere, it doesn't make the formula wrong.
We are not discussing the formula. But the formulation of the problem should have been clarified.
The full field of events was written out earlier (for association).
Now, if you want to use the formula - may be you must find out how often these indices generate signals?
;)
I wonder how to account for the correlation of indicators. All indicators depend on the price, just the processing algorithm may be different in each one.
But the correlation is not going anywhere.
This is probably a different topic. Correlation takes into account the consistency of the processes, and the above Bayesian formulas give a static integral snapshot of the mutual behaviour of the two indicators.
In principle, dependence on the same variable (price) does not mean correlation.
Didn't you, Simon, say the other day:
?Yes, I did. But there's no contradiction here, nor is there any other statement... you can't make all the money, but that's no reason not to work... you can't clean up all the rubbish, but that's no reason not to clean up...
.
.
MERRY CHRISTMAS!
.
.
да, говорил. но противоречия здесь нет. как и в других высказываниях... всех денег не заработать, но это не повод - не работать... весь мусор не убрать, но это не повод - не делать уборку...
.
С РОЖДЕСТВОМ!!!
.
I have not quoted here to show the contradictions, nor to rebuke them. People here have spoken out, each as best they can. All opinions are valuable. Those who wish to find the truth will have to spit out the husk from time to time.
> MERRY CHRISTMAS!!!
Likewise!
avatar in the thread absolutely rightly questioned the ideology of decision making when combining indusers, asking a direct question about simultaneously firing guns. I gave the same answer there, which seems to argue in favour of the fact that it is more profitable to multiply indusers than not. Contradiction.
We have a different logic here than with simultaneous firing of guns on target.
Traders believe that in order to make a decision using two indicators, each of them must show the same direction, not just one. Generally speaking, this premise is optional.
For a complete, exhaustive consideration of our situation we would have to consider the probabilities of a much larger number of combinations of signals. For example, consider an exotic one like (A+ OR B-).
If we assume that the market can go only in two directions (up or down), then similar conditions, the a posteriori probabilities of which will have to be counted, will be not two, but 16 (again, if the logical bundle can be two - AND or OR). No one forbids you to calculate them. But you can't do without Bayes formulas. Here's the beginning of a complete list:
M+ | (A+ OR C-)
M- | (A+ OR C-)
M+ | (A+ AND B-)
M- | (A+ AND B-)
etc.
A posteriori probabilities may only be needed because these indicators fire at different frequencies.
I have said this before.
Otherwise, just write out a complete ring of events.
Well, there's the answer.
I'm not going to write out a full ring here. Whoever needs it, will write it out. But the topic, given this complication in the logic of decision-making, is already becoming very curious.