Steady EA - page 2

New comment
 
pooh123:
Hi elitecamper,

I coded an EA based on the same system from Person too. Basically it's 100% of my understanding of what Person described for the system. I posted it in Request & Ideas. Backtesting shows that it's not profitable, although it has a winning % of 69% during 10 years backtest, due to the unfavorable tp and sl ratio of 40 vs 100. The test report in the book shows the system written in Easy Language has a winning % of 67%, but is hugely profitable. So I'm puzzled. Now I'm glad to see your are working on the same system. It's like I have a comrade now.

I've put your Steady-EA on forward testing since Thursday evening. However, it didn't place any trades. There're no error messages either. How often does this EA place trades ?

One other thing, I checked the codes and it seems to me that, only S1 and R1 are used for signaling. Have you tried S2, R2,S3 and R3 and discarded them for being less reliable ?

Yeah I would say that I think this author embellishes a little. In regards to the Steady EA it doesn't trade often, I backtested it for the year 2011 on Euro and it did 80 trades. Also I only used S1 and R1 in the codes as I didn't get to the other ones. I am not sure if the other ones would be profitable or not. Also I would say that this EA looks pretty good in the back test I posted, but I have tested it in other years and it has losses. I need to code a new one that can bare some draw down. Right now if it goes into draw down it may close the trade instead of waiting it out so that needs to be fixed.

 

Just did a new version of the pivot EA. Elitecamper, hope you won't mind that I post it in your thread. In Person's original system, one condition was for both fast and slow line to move out of overbought oversold area, in this version I removed the condition for slow line. Only need fast line to move out of overbout oversold. So entry can be at least one bar sooner.

The 2nd change I made was to use Mladen's indicator Cronex Taichi as a filter for trend. Only send long signal when Taichi is above signal line. Only send short signal when Taichi is below signal line.

Those two changes seem to work. A 10 year backtest show the system is now profitable. But, modelling quality is poor and a pivot based EA is a MTF EA so we cannot trust the backtest too much. Anyway, it can be a start and I'm working improvement.

 

Hi pooh123,

Is this ea still for 15 min time frame?

Thanks.

pooh123:
Just did a new version of the pivot EA. Elitecamper, hope you won't mind that I post it in your thread. In Person's original system, one condition was for both fast and slow line to move out of overbought oversold area, in this version I removed the condition for slow line. Only need fast line to move out of overbout oversold. So entry can be at least one bar sooner.

The 2nd change I made was to use Mladen's indicator Cronex Taichi as a filter for trend. Only send long signal when Taichi is above signal line. Only send short signal when Taichi is below signal line.

Those two changes seem to work. A 10 year backtest show the system is now profitable. But, modelling quality is poor and a pivot based EA is a MTF EA so we cannot trust the backtest too much. Anyway, it can be a start and I'm working improvement.
 
stanley18:
Hi pooh123,

Is this ea still for 15 min time frame?

Thanks.

Stanley,

Sorry I'm late with the reply. Didn't check in for quite some time. Yes, this EA is for m15.

pooh

 
poohguo:

Stanley,

Sorry I'm late with the reply. Didn't check in for quite some time. Yes, this EA is for m15.

pooh

Hi, Pooh, i know this might be a long time after the post been discussed, however I am interested in writing code in mt4, and I also like to apply pivot point and stochastic, so I found your code and tried but. 

Can you tell me what is the 'P_distance' variable purpose of use ? cause I couldn't figure out and the EA seems no action, which I am still trying to find out, hope you can guide me if possible. Thank you again. 


Sean 

12
New comment