Topics
Replies
galafrin
22 Apr 2016, 22:29
( Updated at: 21 Dec 2023, 09:20 )
I didn't hear about publicly available market scanner but it is no big deal to code as it needs only a list of symbol to be filled as there is not yet built-in one.Here is a screen shot of a market scanner that displays highest HL by asset class::
@galafrin
galafrin
16 Jun 2015, 23:55
Last time I checked out, Calgo would not provide leverage, lot size, nor margin required of a symbol, so we have to do it ourselves.
We need a function that returns the leverage of a symbol, then a function that returns the conversion rate from one currency to another, then a function that returns the lot size of a symbol. Those are used to compute the margin required by one lot of a symbol. The max volume will be free margin divided by margin required in lot units.
Margin required = Lotsize * symbol bid * rate (quote currency to account currency) / leverage
max volume = free margin / margin required
@galafrin
galafrin
20 Oct 2014, 23:58
RE: RE:
galafrin said:
Spotware said:
Genetic algorythm uses to reduce drasticaly the number of passes, but in this optimizing it seems not that has happened-
It took 1 h 5 m for a Bar expert optimization of 3 month H1 and 765 passes browsing all the passes posibles, that is brute force in place of GA , while I expected around 200 passes and 15mn max execution, based on other optimizer. The result seems correct.
Genetic Algorithms (GA) could vary between different applications. We plan to expose GA parameters such as: population size, minimal iterations count, stagnation period, etc. After that you will be able to tune GA to be faster.
For "Bar expert" optimization you can try m5 (open prices) data mode. It should speed-up optimization process.
Indeed, by default the data mode is set to tick data so it looks like the optimizer worked with ticks even though on H1 bar expert. Anyway the elapsed time has fallen to 48 s , for exhaustive parameter browsing, much less than expected.
And it does reduces passes a lot in some configurations from about 350 times , specially when the parameter range is big it seems.
@galafrin
galafrin
20 Oct 2014, 20:21
RE:
Spotware said:
Genetic algorythm uses to reduce drasticaly the number of passes, but in this optimizing it seems not that has happened-
It took 1 h 5 m for a Bar expert optimization of 3 month H1 and 765 passes browsing all the passes posibles, that is brute force in place of GA , while I expected around 200 passes and 15mn max execution, based on other optimizer. The result seems correct.
Genetic Algorithms (GA) could vary between different applications. We plan to expose GA parameters such as: population size, minimal iterations count, stagnation period, etc. After that you will be able to tune GA to be faster.
For "Bar expert" optimization you can try m5 (open prices) data mode. It should speed-up optimization process.
Indeed, by default the data mode is set to tick data so it looks like the optimizer worked with ticks even though on H1 bar expert. Anyway the elapsed time has fallen to 48 s , for exhaustive parameter browsing, much less than expected.
@galafrin
galafrin
19 Oct 2014, 20:56
( Updated at: 21 Dec 2023, 09:20 )
Genetic algorythm uses to reduce drasticaly the number of passes, but in this optimizing it seems not that has happened-
It took 1 h 5 m for a Bar expert optimization of 3 month H1 and 765 passes browsing all the passes posibles, that is brute force in place of GA , while I expected around 200 passes and 15mn max execution, based on other optimizer. The result seems correct.
@galafrin
galafrin
18 Oct 2014, 23:21
RE:
lec0456 said:
no, no. I mean when you check or uncheck the output to be displayed, like %K in the window above: if there a way to read that variable within the code?
Like
if (!PercentK.Hidden){ ChartObjects.DrawText("STOlabel1", DisplayK, index, PercentK[index], VerticalAlignment.Center, HorizontalAlignment.Right, Colors.Green);)
In my understanding Calgo does not provide yet access to any user chart settings , like indicator box or graph (candles, bars, etc)
@galafrin
galafrin
03 Jun 2014, 22:51
Spotware may relase script capabitity to let users code their own optimizer and access Ctrader primitives. Anyway coding an optimizer is no big deal , specially if focus is put on pips result, change percent and leverage, a few easy basic indictor would do. The point is acces to prices.
@galafrin
galafrin
03 Jun 2014, 22:27
I don't experience any crashes nor CPU overload nor frequent feed problems since a lot of time specially after getting rid of Windows 7 and running windows 8.1, and before I only had a few at times. I used to reinstall windows 7 frequently when it get clogged, also poorly coded tier indicators could hamper the runtime.
Beside, MT4 has been coded for Windows XP, it is no match and will never be for Dot Net software like Ctrade and Calgo.
@galafrin
galafrin
28 Jan 2013, 00:44
Giving that short were not implemented in the preceding version according to the report, here there might be dissimetry between short and long trade decision because average long profits roughtly match average long losses while average short profits triplicate average short losses, so there might be room for further improvement toward 430k net profits ;) I don't remember having seen such a difference in yield between shorts and longs when decisions are symetric and the number of bars is huge.
As average short profits are about 100 pips , max short profits about 250 pips, max short losses about 75 pips, the robot might have a 75 stop loss and a trailing stop logic of about 50 pips . It could add position every 15mn while in the trend , that might explain the losing streak at 410.
@galafrin
galafrin
25 Apr 2016, 00:26
Did you try the converter from MT4 to cTrader ; http://2calgo.com ?
@galafrin