Macd Stochastic Fractal MA5/25 Daily Marylin

[niepce]

I've crawled thru 5 or 6 fractals indicators via google.
None of them had a custom deviation setting.. If you know one, just pm me.

[jjgengis]

Hi

in these holiday I'll try to find a workaround to solve the problem. I'm not a coder, but I think that something I can do.

Happy Easter to all!



JJ

[PhP71]

hi all,

having the same issue of high CPU usage, I dig a bit in the code (Marilyn and SS_indi) and understood why.

Phil, like you said the problem is the call to the icustom routine.

As Marilyn is calling SS_indi only to get the last fractals (high/low, fast/slow), I transfert the fractals code from SS_indi the Marilyn and compute them only on new PERIOD_FRACTAL bar.
I hope I didn't transgress any rights by doing so.

Nothing have been removed from the code of Marilyn, just the replaced parts commented out.
We can now backtest (on one pair at a time and for what it worth) to optimize the settings without waiting ages.

Have fun.

[niepce]

I hope I didn't transgress any rights by doing so.

C'mon you ain't serious


Thanks a lot for contributing !!!

[AGT]

hi niepce ! as I understand its now even better to use PhP71's version for regular trading too ... instead of yours ? His mod is not only good for BT's isn't it ? May we use the setfiles posted earlier for the recommended pairs without any changes, that means: can we easy copy them into the PhP71's "no_indi" bot ?

[niepce]

Yep, just run this one, install your 0.35 setfiles, make sure about RSI values as I explained earlier, make sure you manually put the SS indicator on your chart, and that's it.
If your magic number is identical, MM will take over your running trades.

--
Regarding the "Is Backtestable" statement : I have a doubt about the RSI and the Stoch_H4 routines - since they rely on different TF the EA is running on.

[niepce]

Just made a backtest with my "shit for M1" data, and I don't like what I see.
Anyone with quality M1 can post some BT ?

[phil_trade]

Just made a backtest with my "shit for M1" data, and I don't like what I see.
Anyone with quality M1 can post some BT ?

You first need to verify in visual mode if Marylin perfectly follow all the rules, if no logic error etc

Check every trade and every close, before M1 quality data, as without debugger in MQL, it's very difficult to avoid them.

happy we

Philippe

[theforexedge]

Just made a backtest with my "shit for M1" data, and I don't like what I see.
Anyone with quality M1 can post some BT ?

niepce

Here is a good source of 1M data:

http://www.forextester.com/data/datasources.html

Not all pairs but most of the majors

JP

[phil_trade]

Hi all

a very important parameter DeltaPipsMA25= ? ( Delta min in pips between today and yesterday EMA to allow trade) is not enough explain in Niepce Empty4 version :

0 mean Conservative mode
5 or 10 means Aggressive mode

example with FT backtest for €/$
with 10 you will have more bad trades / DD but better result in overall period
with 0 no trade against MA25 so less bad trades but lower overall result as you catch the trend later.

live axample this morning with USDCHF with 5 pips -> New SELL trade USDCHF this morning.
It would not happen with DeltaPipsMA25 = 0

You could also test -5 to force a well established MA25 trend...but you will be late in the trend

Philippe