Counsellors to whom. Lots of them and for free! - page 12
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
Dear Yuri, how do I get "access" to the strategy/advisor repository? I had the PRS reload the terminal many times after "creating" a strategy, but it didn't save anything...
By the way, do I need to optimise this EA and what method should I use?
If you know how to optimise it, you should try it.
But judging by your own questions, it won't do any good.
P.S. By the way, how often (every week/month) do you need to generate (or optimise an existing strategy) to adapt it to a changing market?
Well, if you need 10 new strategies every day, you can generate and optimise them every day.
If you want to be a normal trader then 2 or 3 is more than enough for a long period of time (until profitability starts to deteriorate noticeably).
Thanks! I figured it out.... just didn't get it right away ) first time SSB turned off after the 1st suggested and saved strategy... Spring must be doing something to IQ ((
No surprise there. The bulk of the SSB 'inoperability' is the user's own doing. And a small part may be due to technical reasons, such as broken internet connections.
Although the program interface is made in such a way as to reduce this very amateurishness to a minimum, ie, only two lists and one button. And yet, even with such a primitive interface, some particularly advanced individuals manage to think out and do that or that.
The resulting EA has 11 parameters that affect entry/exit. How to optimise ideally?
1.Parameter range (I use 1 to 100, step 1).
2.Date range (I use 2008.01.01 - today)
3.Thickness tolerance test (I uncheck "use date" and test it on the whole period 1999-2009)
4. Selection of results (use 70% of profit, 30% of loss; more trades - the more the better)
5. Optimize all variables together or in groups (I try to optimize all of them together)
The resulting EA has 11 parameters that affect entry/exit. How to optimise it ideally?
1.Parameter range (I use 1 to 100, step 1).
2.Date range (I use 2008.01.01 - today)
3.Thickness tolerance test (I uncheck "use date" and test it on the whole period 1999-2009)
4. Selection of results (use 70% of profit, 30% of loss; more trades - the more the better)
5. Optimize all variables together or in groups (I try to optimize all of them together)
The best solution is not to do additional fitting of EAs to the history - in most cases it will be a waste of time, but to run additional tests using default input parameters. I.e. forwards are mandatory, because SSB repository is less than a week old, and there is nothing that has passed any kind of test. All the strategies in the repository have only been run on different timeframes and symbols. And the rest of the tests by R. Pardo would not hurt.
Well, this picture is exactly the opposite. The red bars are downtrends. The value 0 - entry into a downtrend. (when scalping).In addition 5 original signals to the sell (yellow line cuts off). And 2 losing buy signals in the beginning.
After a closer look at the oscilloscope readings, the picture turned out to be not so bright, as it was at first sight. Of course, there is some clarity but interpretation of intermediate signals is not clear. The Expert Advisor gives basic signals without the oscilloscope.
It turns out that signals interpretations for each strategy are different and there are no universal methods. I.e. of course we can take a strategy, get oscilloscope readings and formulate interpretations. But it takes a lot of time and as a result, until we obtain a more or less adequate interpretation, the market has changed and we need to start from scratch.
It is a pity of course. But I got another proof that not everything is shining, till you look closer.
I think it is better to generate strategies and Expert Advisors without oscilloscopes so as not to fool ourselves and others.
I studied oscilloscope data more attentively and the picture was not as bright as it seemed at first sight. Of course there is some clarity, but interpretation of intermediate signals is ambiguous. The Expert Advisor gives basic signals without the oscilloscope.
It turns out that interpretations of signals are different for each strategy and there are no universal methods. I.e. of course we can take a strategy, take oscilloscope readings and start formulating interpretations. But it takes a lot of time and as a result, until we obtain a more or less adequate interpretation, the market has changed and we need to start from scratch.
It is a pity of course. But once again I am convinced that not everything is shiny until you look closer.
I think it is better to generate strategies and Expert Advisors without oscilloscopes so as not to cause much trouble for ourselves and others.
Well, firstly, the oscillator provides visualization of the strategy, which is a plus.
When the Expert Advisor gives signals, unfortunately, without it you can not see it and analyzing the work blindly is quite difficult, and this way you can review the history.
In addition, I have divided the oscillator EAs into 2 types, one of which is less reliable because despite good performance on historical data it can give out a good los because the system is on breakout and buys at the maximum price, after which the price usually rolls back.
Well, secondly, the interpretations for different types of strategies are different (but there are only 2), but each has its own discretization under certain market conditions. Of course it's hard to program, but it can be successfully used for entry when scalping the market, it's harder with exit.
Maybe some people should not study these issues in detail,
they should blindly trust the "black boxes". The idea of scalping in general is individual for everyone and even for me it was just a confirmation signal. But I was surprised by how strong and specific this signal is, because it is optimized by the opening prices and as soon as a new bar and signal appear, it starts to turn into reality.
Well, first of all, the oscillator gives visualisation of the strategy, which is already a plus.
When the Expert Advisor gives signals, unfortunately, without it, it is not visible, and it is difficult to analyze the work blindly, and so you can review the history.
In addition, I divided the oscillator advisors into 2 types, one of which is less reliable because despite the good performance on the historical data, it can give a good loss, because the system is on the breakdown and buys at the maximum price, after which the price tend to rollback.
Well, secondly, the interpretations for different types of strategies are different (but there are only 2), but each has its own discretization under certain market conditions. Of course this is hard to program, but can be used successfully for entry when scalping the market, it is harder with exit.
Maybe some people should not study these issues in detail,
they should blindly trust the "black boxes". The idea of scalping in general is individual for everyone and even for me it was just a confirmation signal. But I was surprised by how strong and specific this signal is, because it is optimized by the opening prices and as soon as a new bar and signal appear, it starts to turn into reality.
The black box is when the strategy is closed. In this case there are no such trade secrets.
Even deaf black boxes can be driven by various tests to check for lousiness.
At least I have never been able to formulate any meaningful interpretation of the oscilloscope. And the forward tests are much faster and clearer in identifying and filtering out the fittings. That is why I prefer the Pardo method, already proven in practice, rather than subjective and ambiguous interpretations of oscillations.