Algorithmic ''centrifuge'' - page 7
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
Hello Peter!
I think you can run the script through the history extremums and collect the statistics of the values the indicator took at that moment. Most likely we will get such a dinosaur:
There are two questions:
- how do we get extrema in the "peeked" history?
- what indicator should be used? After all, a simple, price-derived indicator will sometimes be resold and repurchased for a long time.
A mixture of several indicators or time-varying parameters of one indicator will surely provide a fit for the history. As people have already written here. To try to identify general patterns instead of memorizing all particular ones, the input parameters should not be enough. And you know it yourself :)
In general, the question is not about the indicator that moves up and down behind the price, but it sees the trend, knows the current location, feels the levels, uses statistics and so on.
Greetings.
You can make a special algorithm that uses tester Optimization to find ideal entry points.
After that, on these points you can select indicators for composing a trading signal of the future strategy.
The principle of selection - as close as possible the repetition of an indicator (or an important indicator from any custom formula) at each entry point and each exit point.
The resulting indicators/formulas, as a result of "natural selection", will fit into the series of Trade Signal parameters in the Component TS.
Even so, it doesn't matter.
A tester is not necessary.
The maximum number of indicators in an effective EA is one, but zero is better. You will realise this when you have played with optimisation. The sooner it happens, the better for you, but apparently, you have to go through with it.
It's great, that your field of interest has shifted to the area, for whichmqlis intended . Congratulations!
do you have a solution for this problem? - we are standing at an arbitrary point on the chart (not terminal! :) ). buy now or sell?
it's not as obvious as it seems, even if you see the future - take 300 pips in a day or 1500 pips in a week? (this is just an example, days can be replaced by hours and hundreds of pips by tens)... it's the so calledshoreline problem, only instead of segments to find the length of the shoreline there will be alternating sell/buy trades.
an unambiguous answer to this question would be the exception rather than the rule. and without it, no automatic search would work...
do you have a solution for this problem? - we are standing at an arbitrary point on the chart (not terminal! :) ). buy now or sell?
it's not as obvious as it seems, even if you see the future - take 300 pips in a day or 1500 in a week? (this is just an example, days can be replaced by hours and hundreds of pips by tens)... it's the so called shoreline problem, only instead of segments to find the length of the shoreline there will be alternating sell/buy trades.
An unambiguous answer to this question would be the exception rather than the rule. and without it, no automatic search would work...
You are wrong. Having a tick history in the Tester, you can use the Optimization mechanism to find the IDEAL entry and exit points. You can adapt the Genetic Algorithm to find these points. Having the points, it is possible to find appropriate indicators. And all this - to perform automatically.
Exciting, isn't it?))
You are wrong. With a tick history in the Tester, you can use the Optimisation mechanism to find the IDEAL entry and exit points. You can adapt the Genetic Algorithm to find these points. Having the points, it is possible to find appropriate indicators. And all this - to perform automatically.
Exciting, isn't it?))
you forgot to capitalise the Entry and Exit Points...and the Mechanism of course too
Without it, it's not very interesting at all :-)
you forgot to capitalize entry and exit points...and the mechanism of course also
without it, it's not shitty at all :-)
There are a lot of things you forgot to write here.
I've already asked you to draw a block diagram, because it is unclear where we are going in this guillotine
here are the latest messages from the topicstarter, that let's look for the ideal entry-exit points through the optimizer, to strive for these points later with an automatic selection of indicators..... although you can unload the ZigZag and do something with it. What to do is not clear.... i do not understand how to connect indicators, how to identify the entry/exit in the search for TP - in general, i do not smell automated
...
You are demanding a ready-made solution and schemes, while the concept is being discussed. Not yet.
I invite others to think too and propose their own methods of automating the assembly of strategies.
So far only one participant is making concrete suggestions.
If you have any ideas how to use GA for searching for ideal entry points on the history, speak up. I have not decided yet.
If you have any thoughts on how to use GAs to find ideal entry points on history, speak up. I have not decided yet.
GA is not needed, moreover it does not know how to do it
You should upload the ZigZag to a file and load this file in its entirety when you run EA in the structure array, I did so, in the optimizer without problems all works
UPD: I remember, do not upload it to the structures array, but load it to the int array and make the direction of trade positive or negative int
.... it is not clear at all how to connect indicators, how to determine the input/output when searching for TC - in general it does not smell of automation at all
The indicators will be included in the general program as callable functions. Each indicator will be represented by one parameter and its value at entry and exit points will be written in an array. At the end, it will be possible to categorise the indicators by value. The closer the repetition of the values at the ideal entry points, the more accurate the indicator.
1. GA is not needed, moreover it is not able to do it
2. unload the ZigZag into a file and load this file in its entirety when you run EA in an array of structures, I did so, in the optimizer without problems all works
1. GA can be adapted to narrow down the search for ideal entry points.
2. ZigZag will not show perfect entry points. That's not it. There will be a large margin of error there. Optimizer with GA can do much better. IMHO.