How to become a successful Forex trader with algo trading


Hello, dear traders, this is Petko
Alexandrov and I continue now with the reactor. This is the tool you can find, if
you click the arrow next to the generator, there is the reactor, there is
the validator. So the validator is where you can drop old strategies, collections,
or complete folders with strategies. And you can validate them with a new
historical data or with data from different broker. And it’s very nice
because when we generate strategies, we save them in collections and after that
we can validate them. So we don’t lose time again to generate new strategies.
But we can just filter them through the validator with the recent historical
data. So the reactor is the tool where we do everything automatically. And we don’t
need to perform robustness testing manually, or Walk Forward optimization
and validation. It is done automatically by the software and by the reactor. So
first, we have the generator settings. OK? I will demonstrate it to you how it
works and I will stick here to the GBPJPY on M15 chart. As strategy
properties, I will keep it the same 0.01 as the minimum lot. And
then I will use range from 10 to 100 pips for the Stop Loss and same for the
Take Profit. And as well in generator settings, I will
keep it search best as Net balance so I will see the strategies with most
balance on the top. And here I have the chance to choose Out of Sample and In
Sample. But here I will stay with In Sample. So I will be using the Out of
Sample in the Walk Forward optimization. And below is the common Acceptance
criteria. If I click on it, you will see that I have minimum backtest quality of
98 at the moment, Minimum count of trades 300, which is just fine. And I will add
minimum Profit factor of 1.1. Where is it, right over here, 1.1.
And as well I will add minimum Profit factor for the In Sample part and for
the Out of Sample part which I will use in the Walk Forward optimization. OK? Profit
factor 1.1 and here as well i will choose Profit factor of 1.1. So I want to
have Profit factor of 1.1 for the complete backtest for the In Sample
part and for the Out of Sample part. Now, going back to the reactor, I will go to
the Walk Forward optimization. I have the choice whether to use it or not. Actually,
all the tools below I can select if I want to use them or
not. At the current moment, I have selected the Walk Forward optimization
and the Monte Carlo validation and this is what I will use now, and you can see
they are displayed on the right side. So what is the idea of the reactor? That
when the strategies are being generated they automatically go to the
Walk Forward optimization or the tool you have selected here to use. And if
they pass this optimization, or the validation, or whatever you select to use,
they will go to the next one. So from the Walk Forward optimization, in this case,
they will go to the Monte Carlo validation. And it will perform the
variety of tests that I have shown in the previous lecture. And if the strategies
pass, then I will see them in the collection. OK? So here, the reactor does
all the job automatically, and before we were doing it manually. For each strategy
we were performing optimization, robustness testing, and now when the
reactor is available, it’s available actually from last year, it is very easy
to do it because we just need to set up the reactor and it filters the
strategies automatically. And at the end, in collection we have the strategies
that are within our Acceptance criteria. So for Walk Forward
optimization, what I will be using is a lower number of segments. So I will go
down to 3. So it will be the very same thing that I have already showed
but with 3 segments. My recent tests showed that it works a little bit
faster to generate strategies and personally I like it at the moment to
work with 3 segments and not with five. Of course, if you want to generate
your own strategies, you can test with 3, with 4, with 5, with any other
number. OK? For Out of Sample, I will stay with 30%, search best; net balance, and let
me have a look at the settings that I have in the Walk Forward. In settings I
will choose 3 segments, Out of Sample below is just fine, as a validation I
will stay with minimum of $10, and here are the parameters, the statistics. And
actually here you just have a look at the statistic. Alright? Going back to
the reactor, I have the Out of Sample percentage, Net balance, the Walk Forward
settings are fine, validated segments if I choose to use 3 then I want to
have 3 out of the 3 segments validated. And here you can select if you
want to use the common Acceptance criteria or not. Yes I want the final
backtest with the last parameters to be validated with the Acceptance criteria.
And I have already said it in the generator, it’s the same here. So I would
like my strategies to have 1.1 Profit factor for the complete backtest
for the training part and for the trading part. OK? And then we go to the
Monte Carlo, I will stick here with the default settings. And as I have shown, I
use normally the strategy variations, the randomized indicator parameters, and the
randomize backtest starting bar. So this is how I set up the reactor, and this of
course, is an example. One more time, if you want to do it in a
different way you are very welcome to experiment. This is how actually traders
learn trading. OK? I will click on start and you will see the strategies are
being calculated. So one more time, those of the strategies that pass the
generator, they will go through the Walk Forward optimization, and then if they
pass the Walk Forward optimization they will go to the Monte Carlo validation. So
in generator, the strategies need to pass the Acceptance criteria. OK? The very
same Acceptance criteria that is after that with the Walk Forward optimization.
Now, what is the great thing with EA Studio? It is web-based and I can set the
very same thing on another browser. So if I open one more time our website,
EA Forex Axademy, and I go to the Expert Advisor Studio which is integrated and I
log in to my account, I have some old work here with EURUSD strategy that
loads automatically. And I will run now a reactor but I will separate the top. OK?
It works faster if I separate the top. So I have already done it for the GBPJPY,
now what I will do, I will set it up for the EURGBP. I will stick on M15
and then the same settings should be saved but I always check them one more
time. And here, I will use working minutes 600. OK? Let me have a look if I
have done that with the other, no I forgot to do that. So let me do it for
600 minutes. OK? I want to run the reactor overnight, now it’s at the end of
my working day. And I normally run the reactor at the end of the working day so
the morning I have new strategies into the collection. So 600 minutes is exactly
10 hours and it works fine for me to leave 3 reactors working overnight. So I
will do the same thing here, 600 minutes for the EURGBP. The
Acceptance criteria, it did not save. In this case I will need to do it one more
time, so 300 count of trades and minimum Profit factor of 1.1. And I will do the
same for the Acceptance criteria in the In Sample part,
minimum Profit factor of 1.1. And then I will have a minimum Profit factor of 1.1
as well in the Out of Sample part. OK? Here it is. And then for the reactor, I
will choose to use Walk Forward optimization which is right over here.
And of course, I need to select 3 as well as validated segments. Let me have a look
at the Walk Forward settings, here should be 3. And here it is, it’s ready now. And
the Monte Carlo is saved to 20 and 80%, same for the strategy variations. OK? So I
think I am ready now and I will click on start. And you can see both
reactors work simultaneously. OK? This saves a lot of time because we can run a
couple of reactors. I normally run 3 to 4. Just don’t make it too complicated,
make sure you organize your work if you decide to generate your own strategies.
So one more time, I will open the Expert Advisor Studio, separate tab. And once you
are used to do the setup and you are used to work with the Expert Advisor
Builder, you will see how quickly the things happen. So I will have it for EURJPY,
right? Let me check one more time, here I have it for EURGBP, and here I
have it for GBPJPY, and I will have it here for the EURJPY. On H1 chart,
strategy properties, it is the same. Minimum lot of 0.01, Stop Loss range from
10 to 100, Take Profit range from 10 to 100, and generator settings 600 minutes.
Common Acceptance criteria, let me do it, so it’s 1.1
Profit factor, 300 count of trades, as well Profit factor of 1.1 for the
In Sample part and 1.1 for the Out of Sample part. OK? Here it is. Going
back to the reactor, I will select the Walk Forward optimization where I said
3 segments out of the 3 should be validated. And as well in settings I will
change to 3 segments. OK? And now I am already with the last reactor. I will
click on start and you will see now that 3 of them are working simultaneously,
and let me bring them together on the screen. Here it is, it’s for EURGBP
on M15, GBPJPY on M15 and EURJPY. Actually here I did not change the
period, it’s on H1, I wanted to be on M15. Now, it is ready and you can see 3
of them are working simultaneously on one computer. And if you run more, they
might get a little bit slower. And at the moment I have another 3, or actually 4
open which I started earlier. That’s why it’s getting a little bit slower. But
normally, I keep it 3 to 4 just to take the best of it. And you can see here
that for the EURGBP, already one strategy passed the Acceptance criteria
in the generator, or this is the common Acceptance criteria. Right? Then it went
to the Walk Forward optimization, it passed the Walk Forward optimization. But then
it couldn’t pass the Monte Carlo. OK, guys? So this is how it is working, it is very
easy to set it up. I will leave it now and I will show you what are the results
that we generate. Thank you for watching. Cheers.

Leave a Reply

Your email address will not be published. Required fields are marked *