Sidus_EA_wRSI & MA

[traderduke]

Fellow pip chasers
I believe I have a good indicator and I’ve put it into an EA. The EA is a rather simplistic and a little clumsy not as clean as Steve’s. It wins on EURUSD 1H, 4H or Renko charts, OOTB is the settings, BUT MBTrading has lousy Data bases. If some who has access to some decent data could help test it both 4H and Renko. Any hints on getting some reliable data would be appreciated.

Renko familiarity a must! List Renko conversion EA

Let’s use Gamma's EA template for control!


Looking forward to richer days

Ray

[DING]

My EURUSD 2011 H4 OOTB Test Graph.

[skoda2008]

Have you tried this method to generate tick data based renko charts for backtesting?

http://www.forexfactory.com/showthread. ... ost4335984

What size renko bricks do you recommend (or is that what the backtest is for)?

[GarethC]

Hi Traderduke

Have you had a look at constant range bars (CRB) instead of renko? Also could you please explain how the indicator works. I have put it on a 8 CRB GU and EU charts. If I look at only the arrow's it gives some good signals, but also a number of bad ones.

thanks

GAC

[traderduke]

GarethC
I've tried CRB and Renko appears to be the best Offine.
I hope these explainations help. I use
1. 15 pip renko,
2. Sidus 3, 4, & ma 100.

See attached, Hope this helps,

Ray


// Green arrow fast > slow above 50MA; trend = 1
if (diffu > diff_Level && rsi>50 && rsi > rsi1 && ExtMapBuffer1 > ExtMapBuffer5) trend = 1;

// Red arrow fast < slow below 50MA; trend = -1
if (diffd > diff_Level && rsi<50 && rsi < rsi1 && ExtMapBuffer1 < ExtMapBuffer5) trend = -1;

//Long postion exit, trend changing from 1 to 2
if (trend == 1 && ExtMapBuffer1 < ExtMapBuffer2
&& trend[i+1] == 1 && ExtMapBuffer1[i+1] > ExtMapBuffer2[i+1]) trend = 2;

//Short postion exit, trend changing from -1 to -2
if (trend[i] == -1 && ExtMapBuffer1[i] > ExtMapBuffer2[i]
&& trend[i+1] == -1 && ExtMapBuffer1[i+1] < ExtMapBuffer2[i+1]) trend[i] = -2;

Hi Traderduke

Have you had a look at constant range bars (CRB) instead of renko? Also could you please explain how the indicator works. I have put it on a 8 CRB GU and EU charts. If I look at only the arrow's it gives some good signals, but also a number of bad ones.

thanks

GAC

[FX Juju]

Hi TraderDuke...

I ve got some Ducascopy data (Jan 2010-Dec 2011)... I tried your EA with default parameters and 15 pips Renko charts on EURJPY...

Have a look to the results... 1 backtest was very good... the other one very bad... I used a plugin to generate Renko chart... That plugin is able to generate 3 kind of Renko charts... The difference is explained there

http://www.the-forex-strategy.com/index ... in-for-Empty4

The nice backtest was generated using Standard Renko charts with wicks and the bad one with a renko chart with wicks and continuous price tracing, which is supposed to be specially designed for backtest...

I confess I don t understand very well why the special renko is needed for backtesting (I haven t designed many Renko EA yet... so no need to backtest so far)

One thing I know is that once I got outstanding backtesting results with an EA of mine, working with range bar charts... But those results were not reliable because my script for range bars charts did not generate good fst files from historic data (worked properly only with live price feed)...

Anyway, I will run your EA from a Renko chart on a demo account... that costs nothing... I ll also ask the opinion of the programmer of my plugin...

Cheers...

[FX Juju]

Ok... I got the difference between the two kinds of Renko...
In fact, in regular Renko, when you have a trend reversal say from bullish to bearish, Open[0]=Close[1], but in reality, the price moves gradually from Close[1] to Open[0]... So there is a wick... and the wick can even go higher than Close[1]...

I m not 100% sure, but I have the feeling that what you see as Open and Close are not what you EA will actually call...

Well... gotta think about it to make it clear... and see how one should call the variables from an EA

[FX Juju]

Well... but at the same time, the EA should call what appears on the chart right, and not what happens between printing of each Renko bars so I don t know in the end...

[traderduke]

FX Juju
Yes, the OpEN & close are moving targets till it CLOSES. I always run my test with wicks for reallity and visual understanding. Baseball anology; Backtesting at best puts you in the ballpark but getting to your seat is something else. Demo testing gets you to your row and only live will show you to your seat. My MBtrading charts have so much data missing I'm lucky if I can find the the ball park.

I have two EAs, Sidus & Keltner Channel, that I feel are great but the numbers I get out of backtesting fail in demo. With your database maybe we can get somewhere .

Let me know if I can help!

Well... but at the same time, the EA should call what appears on the chart right, and not what happens between printing of each Renko bars so I don t know in the end...