Optimisation in the Strategy Tester - page 11
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
I would also like to ask about another subtlety: when we optimise, let's say at the weekend, when quotations "stand", and then Monday comes and quotations start running, while we are optimising, does it affect the optimisation results and if so, how much does it affect them? Maybe it is necessary to enter the account number from "spontaneously" so that quotors do not start running on Monday? How to do it correctly?
This is about it.
When I ran in the tester 100% results on weekends and weekdays (with a normal, working spread) are different, the same story in mt4.
They forgot to make a function in MT5 that would allow you to set the spread in options and tests, I think I'm not the only one who would like that, was it really so difficult to do in the program?
They forgot to make a function in MT5 that would allow you to set the spread in options and tests, I think I'm not the only one who would like that, was it really so difficult to do in the program?
I think I'm not the only one who would like that.
It would jump almost like in real life.
I have noticed that it has an effect. I ran the EA on weekdays and ran it with extended spread now, the result is quite different, spread is about 4 pips (four digits) on Eurobuck now. So there is some spread jam on weekend in mt5 too, so I don't want to run it at weekend, because optimisation will not be correct. I can even see it visually in mt4, the Expert Advisor has been optimizing since Monday, the results curve has gone down since the weekend, it shows that spread affects the optimization results, they have become worse.
Are you talking about MetaTrader 5 or MetaTrader 4 now?
Bring the technical reports from MetaTrader 5 - save the reports for the first and second time via "Results - HTML" and zip them here, please. And specify the testing parameters, including the name of the trading server.
Are you talking about MetaTrader 5 or MetaTrader 4?
Please give me MetaTrader 5 technical reports - save the reports for the first and second time via "Results - HTML" and zip them here, please. And specify the testing parameters, including the name of the trading server.
MT4:
Too bad I didn't save the reports yesterday when the spread was 4, I ran one EA with certain open price settings (it has a function that allows doing that) and ran it now when the spread is 1.5-2 the difference is huge, unfortunately I cannot post the report with extended spread as I did not save it, next weekend I need to save it. Alpari-Demo server.
MT5: I have not understood anything at all. I have tried one EA on MT5 yesterday with all ticks (spread is about 4 too) on Alpari-Demo server. It is not pumping, but it is stagnating. I have tried it this morning but the server is not yet functional, i.e. the prices are the same as last night, the quotes are not moving and the time is from Friday and the result is totally different.
The spread in MT4 affects the tests surely (by the way, optimization results that managed to pass during the night "go up"), but I do not understand what is going on in MT5.
If I catch a difference I will post it all with reports and pictures.
Why? The spread is stored in the minute bar.
It will jump almost like in real life.
In what minute bar, at what point in time? Let's say we are testing from 01.01.2010 to today, the first deal was done on 3 Jan 2010. What spread will be considered, the one that is "now" or the one that was on 3 Jan 2010?
In what minute bar, at what point in time? Let's say we are testing from 01.01.2010 to today, the first transaction on 3 January 2010, which spread will be taken into account, the one "now" or the one that was on 3 January 2010?
SeeMQL5 Reference Guide / Accessing timeseries and indicators / Organising data access. Each minute bar in the history has its own set of values.
I've read it, I do not understand anything, it's written for professionals, not for users:)
OK. The bottom line is this. All data comes from the server to the terminal in the form of sparingly packed blocks of minute bars. Data of one-minute bars are stored in special files (yearly). When a user selects a certain timeframe, price data of this timeframe are generated using the data of one-minute bars taken from special files.
Accordingly, if each one-minute bar stores spread data, then we can speak about historical spread data for each symbol-period pair. These historical spread data should be used during testing/optimization. I.e. as of January 3, spreads from January 3 should be used, as of "yesterday" - yesterday's spreads (as far as I remember, using current day data for testing/optimization is not provided).
How exactly a spread is selected for a minute bar, and how exactly this spread is taken into account in testing/optimization - I cannot say, because I'm not interested in such questions.
OK. The bottom line is this. All data comes from the server to the terminal in the form of sparingly packed blocks of minute bars. Data of one-minute bars are stored in special files (yearly). When a user selects a certain timeframe, price data are generated using the data of one-minute bars.
Accordingly, if each minibar stores spread data, then we can speak about historical spread data for each symbol-period pair. These historical spread data should be used during testing/optimization. I.e. as of January 3, spreads from January 3 should be used, as of "yesterday" - yesterday's spreads (as far as I remember, using current day data for testing/optimization is not provided).
How exactly a spread is selected for a minute bar, and how exactly this spread is considered during testing/optimization - I cannot say, because I am not interested in such questions.