I am keen to test Pipetraders system but have problem with the StepMA - Trading Systems

XKR1962 2007.02.10 08:27 #71

Unable to view the StepMA indicators

Hi guy's I'm a total newbie so I apologise now for my ignorance . I am keen to test Pipetraders system but have a problem with the StepMA.

I have installed Pipetraders template I also downloaded his StepMA and can see the Force index chart ok, however I cannot get the StepMA indicators to show as seen in the charts that Pipetrader has been posting.

I spent a good few hours playing with MT yesterday so any help would be greatly appreciated.

Thanks to Pipetrader and everyone else posting to this very interesting board.

MetaTrader 5 web platform Constant variables Williams' Percent Range -

MrPip 2007.02.10 10:23 #72

bug found in EA

I was referred to this thread by a friend of mine.

I downloaded the EAs and looked at the code.

For the safer version I found the following bug which will cause the if statement to always return false.

For Buy

Force3>Force2 && Force2>Force3

For Sell

Force3<Force2 && Force2<Force3

I changed this to

For Buy

Force3>Force2 && Force2>Force1

For Sell

Force3<Force2 && Force2<Force1

Here is a modified version in the hopes it would speed backtesting.

No changes int actual logic. I just moved some code into functions and optimized for speed of backtest.

That was done by studying the rules and modifying the order that the indicators are called.

I also separated the logic for the 3 sets of conditions to enter a trade and the set of conditions to exit a trade.

I usually cheked for Force first because StepMA is slow to calculate.

I also fixed the bug.

I have not yet looked at the riskier version.

When I do I will make the same changes.

Robert

Files:

pipetrader_safer_v2.mq4 17 kb

Return jump For loop Overview of control statements

gtketum 2007.02.10 12:06 #73

looking at this chart, i think force index is giving different values base on your data feed. i am using alpari and my values for FI are completely different and not even close. I am using the 30 min TF for gbp and FI of 13 just like you. the stepMA matches but the FI values are different. this can be a problem for the EA since trades will be triggered in some platforms and not in others. can anybody confirm this??

PipeTrader:
Check for this conditions...!

2 smaller on the Force Index. Close trade.. I think is a good point to close. No ?

If you can put this conditions in EA for sure is more profitable..

PipeTrader.

Average True Range - Chaikin Oscillator - Oscillators Forced test stop: TesterStop

et_phonehome_2 2007.02.10 12:57 #74

gtketum:
looking at this chart, i think force index is giving different values base on your data feed. i am using alpari and my values for FI are completely different and not even close. I am using the 30 min TF for gbp and FI of 13 just like you. the stepMA matches but the FI values are different. this can be a problem for the EA since trades will be triggered in some platforms and not in others. can anybody confirm this??

Your confirmation is true. At some point, the FI will eventually get there to closely correlate with other brokers. I have the same issue with trendlines, and some customized indicators..

et_phonehome_2 2007.02.10 12:58 #75

XKR1962:
Hi guy's I'm a total newbie so I apologise now for my ignorance

. I am keen to test Pipetraders system but have a problem with the StepMA.

I have installed Pipetraders template I also downloaded his StepMA and can see the Force index chart ok, however I cannot get the StepMA indicators to show as seen in the charts that Pipetrader has been posting.

I spent a good few hours playing with MT yesterday so any help would be greatly appreciated.

Thanks to Pipetrader and everyone else posting to this very interesting board.

For StepMA, you must scroll back your charts far enough for it to calculate correctly since it requires many bars previously.

mikep 2007.02.10 14:15 #76

PipeTrader:
Check for this conditions...!

2 smaller on the Force Index. Close trade.. I think is a good point to close. No ?

If you can put this conditions in EA for sure is more profitable..

PipeTrader.

Alright I think I have done it as requested. There are two versions, one called safer and another called riskier. Here's the main differences:

"Safer": Once we get a StepMA arrow, the FI must be over the threshold and in agreement to enter.

"Riskier": The FI does not have to exceed the threshold when a StepMA arrow is given.

I found in the previous versions, multiple trades would be opened/closed in a bar where a lot of movement occured - this could multiply the losing trades considerable. I tried using the suggestion found here for anyone who's interested: https://www.mql5.com/en/forum/173428

I also noticed that in the previous versions orders closed even when they were not supposed to - I tried fixing that by using a differeny template.

Anyway, this is definitely a work-in-progress as I'm getting a crash course in programming.

I've been trying to visually backtest via strategy tester and they seem to work - I haven't been able to get great data so my backtests are only yielding about 40% quality. Please test and try optimizing each...

Thanks again PipeTrader for sharing your system.

Mike

edit: Thanks to MrPip! The riskier version also had that bit a faulty code. I've corrected both and have uploaded here again.

Files:

pipetrader_riskier_v1.mq4 12 kb

pipetrader_safer_v1.mq4 12 kb

Forced test stop: TesterStop Questions about how to Testing indicators

omelette 2007.02.10 21:13 #77

et_phonehome_2:
Your confirmation is true. At some point, the FI will eventually get there to closely correlate with other brokers. I have the same issue with trendlines, and some customized indicators..

....reason being FI uses volume in the calculation and as there's no reason why different brokers volumes will be the same, FI will differ accordingly...

tnh_z 2007.02.10 21:34 #78

bluto:
I was intrigued by PipeTrader's method so I decided to whip up a little EA to test the ideas. The EA deviates a bit from the posted method and doesn't use the Force indicator. The StepMa signals alone are used to trigger orders. I use some dynamic lot sizing based on equity and a trailing stop feature. The EA seems to perform best on an H1 chart with medium-high volatility pairs. I backtested USDJPY, GBPUSD & EURUSD for 2006 using 90% quality data....backtest reports and the EA are in the attached zip file. According to the backtester, if you had started out in Jan. 06 with $10K for each pair, you'd be worth about $8 million by Dec. 2006. (lol

) One characteristic of this EA is that it makes monster profits during trending periods, and essentially breaks even/makes small marginal profits during ranging periods. Anyway, download and forward test. You'll also need the Step_MA indicator PipeTrader posted. Who knows, might be a cool EA.

Thank you for new idea and work.I'll test it in demo this week.

luxinterior 2007.02.10 21:44 #79

bluto - can you up the N_StepMA_1_Email indicator so i can test your EA please?

Lux

omelette 2007.02.10 21:48 #80

luxinterior:
bluto - can you up the N_StepMA_1_Email indicator so i can test your EA please? Lux

It was posted by PipeTrader in his 2'nd or 3'rd post of this thread.........

My simple 20 pips. - page 8