Discussion of article "Self-adapting algorithm (Part IV): Additional functionality and tests" - page 6
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
These are fixed values for example. Almost everything that concerns the market is corrected by itself in the algorithm. There are fixed values, like the number of blocks to analyse, I use the range of 24-32 blocks. But these values are not related to the market, they are parameters of the algorithm itself, in fact it is the accuracy with which it works. That's about it. Now the percentage of overweight is fixed, but it will be modernised, and it is not a critical parameter, I know what it depends on, I can set it by hand. In general, there is something to improve in order to raise the quality of trade to an acceptable level.
Maybe you should have taken the range from 17-29?
What's the difference? The market changes all the time, and the currency pair changes its behaviour all the time.
I can tell you're not a programmer. Am I right?
Or maybe we should have taken the range from 17-29?
What difference does it make? The market changes all the time and the currency pair changes its behaviour all the time.
I can tell you're not a programmer. Am I right?
Because it doesn't matter how many blocks to analyse. The size is decisive. The more blocks there are, the more accurate price quantisation is, the smaller it is, the coarser it is. Blocks are just a conditional representation of the price to make it convenient. And no, 17 will not fit, it is too coarse and these parameters are justified by certain reasons. Every parameter is justified by a reason. I don't make parameters that need to be guessed, each one can be calculated. You can take 1000 very small ones or 10 big ones, it doesn't matter at all.
I see that you have been developing trading strategies since 2008. I also started Forex since 2008. But I am a programmer and not a simple one :)
Now I will tell you a "secret". Any programme does not think of anything. It is necessary to set some initial values and clearly tell it what to do. Otherwise, it does not think of anything itself.
This is the only disadvantage of a trading strategy, when we set some fixed values of input parameters. It does not matter whether it is time, number of bars, some levels or other.
Let me give you one simple example:
Let's say you set Take Profit (TP), at the level of 30 points (300 at 5 zn.), or it was made by your programme, for some reason.
The price reaches the level of 25 or even 29 pips, the TP does not work, the price goes back and finally closes with a minus.
Question: If you have a self-adaptive algorithm, then what can your algorithm do to prevent the order from closing with a minus? After all, Trailing Stop is not a solution.
I see that you have been developing trading strategies since 2008. I also started forex since 2008. But I am a programmer and not a simple one :)
Now I'll tell you a "secret". Any programme does not think of anything. It is necessary to set some initial values and clearly tell it what to do. Otherwise, it does not think of anything itself.
This is the only disadvantage of a trading strategy, when we set some fixed values of input parameters. It does not matter whether it is time, number of bars, some levels or other.
Let me give you one simple example:
Let's say you set Take Profit (TP), at the level of 30 points (300 at 5 zn.), or it was made by your programme, for some reason.
The price reaches the level of 25, or even 29 pips, the TP does not work, the price goes back and finally closes with a minus.
Question: If you have a self-adaptive algorithm, then what can your algorithm do to prevent the order from closing with a minus? After all, Trailing Stop is not a solution.
But I'm a programmer and not an easy one :)
Very dubious statement....
I know how algorithms work, I develop them and there are no secrets for me. You haven't read the last two articles.
No, I haven't. But I saw that you have the number of some blocks, I guess you mean the number of bars?
If there are fixed values in the programme, then every time you change these values, you will get different results.
No, I haven't read it. But I saw that you have the number of some blocks, maybe you mean the number of bars?
If there are fixed values in the programme, then every time you change these values you will get different results.
That's the point, to answer that I would have to write 30 pages of text. Read it and see if it makes sense. And I don't mean bars. And even why I don't use bars, I wrote here: https://www.mql5.com/en/articles/8136 everything in detail, consistently.
That's the point, I'd have to write 30 pages of text to answer. Read it, see if it makes sense. And I don't mean bars. And even why I don't use bars, I wrote here: https://www.mql5.com/en/articles/8136 everything in detail, consistently.
You write that:
Peculiarities of price series discretisation by time intervals and random component
What principle is used to determine the time interval?
All those regularities, which are determined on history, cannot be used in real trading. Since it is impossible to determine when a pattern starts and where it ends.
Development would definitely go faster if there was a specialist who does it quickly, not half of the functionality in a year and then with problems.
I agree, this is the main problem. You either have to wait or overpay for efficiency.
!!!
Choza mushrooms!!!
I've been writing and writing a comment for 2 hours. Sent it. And there it was, and then it was gone. It's gone.
ai um wshockerbyl....