Hi All

I came across the below linked post on babypips.com. I have been testing and adjusting the code but I am unable to get results that are similar to the original poster's. I have a few thoughts on why, but my Zorro programming skills are limited. I have been unable to contact the original poster. So I was hoping the group could help me fix the code.

Link to original post:
http://forums.babypips.com/expert-advisors-automated-trading/47462-my-zorro-adventure.html

I think there might be two problems. The first is that the code is for an earlier version of Zorro and some syntax needs to be updated. The second is that it seems to be very sensitive to the WFO parameters. So I need to understand how to set these both for testing and ongoing development and trading. I assume that since time moves forward and new data is downloaded, the WFO start date should be fixed. However the cycles will get longer if number of cycles is fixed. hmmmm uncertain about this in practice.

I had a go at fixing it, but I think it is best to get comments based on the original code. The original poster worked with EURUSD. If successful, I will add the loops to cover more symbols, and position size code, and repost.

Thanks

I set the end date to be about the same time as the original post, 28-09-2012. Then using the same number of years and WFO cycles I trained and tested the script. Attached is the equity chart I got and the original posters chart. Despite the code being the same, it is clear that Zorro is trading differently. You can see this during the period around the start of 2010. The OPs chart shows a group of trades held until near the end of the trend in about June. My charts exits them all about a month earlier. Is anyone aware of changes to Zorro that would make such a difference? Is there a mode I need to set?

Original:


My chart:


I changed the series variable declarations to reflect the manual, i.e. vars VariableName rather than var *VariableName. This had no affect, so I assume either is correct syntax. Also I changed the mode line to just set(PARAMETERS); I have also tried training with TICKS flag set. No difference, just took longer.

I just ran the script with the BarPeriod set to 241 instead of 240, just for interest. This is the chart I got. Much healthier looking. Ok....I will wait for comments now.

Look here

http://www.opserver.de/ubb7/ubbthreads.php?ubb=showflat&Number=430277#Post430277

Oh...thanks. I had no idea this had already been discussed extensively on this forum.

Hi DMB,

did you make any progress when using the Weekend parameter?

I was able to find the Weekend setting (=1) that replicated the OPs result. I then looked at other symbols and found it didn't work at all. I was able to contact the OP, who is also a member of this forum. He confirmed that he was unable to get it to work on any other symbol besides EURUSD.

Interesting situation. Since the equity curve has a lot of flat periods, I hoped a portfolio would smooth it out. I haven't tried to figure out if there is some adjustment that will make it profitable on more than the EURUSD. That will be a side project.

Thanks for the update DMB. What a pity that it won't work on other currency pairs.

Yeah. I will look at optimising other inputs and try different exit strategies. But first, is there a signal processing equivalent for the RSI indicator? I get the feeling that HighPass is not equivalent, but I don't know much about the signal processing functions.

I can't answer your question. Sorry. But I can take a look at it. But there are other indicator which will show you when the market is overbought /oversold ...