Live trades for NN EA idea - page 4

 
jdpnz:
Trading with NN's for some 7 years now, I thought perhaps some of my ideas may be of interest here.

I completely agree with the earier post of Project1972 - that curve is just about what one would expect - well, in my experience it is.

However - there is a way to overcome the out of sample data flat line which I find quite successful namely to, have an array of NN's and a catagorizer in front, deciding which NN to use in any specific case.

It is like - well, to use the army example of earlier, if I have a problem between cloudy/sunny days, it would mean that I need two NN's - one for cloudy and one for sunny days. I also need a catagorizer to decide which one to use and then (well, with market price at least) I find some interesting results.

Also - using a single NN the result is like a (very nice) curve fit but as soon as the market pattern change, the pattern is not applicable anymore.

However, when I implement a structure like the above, the output seems to be much more accurate in following the market.

I think we can also see this when Ehlers says that (for a short period of time) the market may follow a certain pattern - then the pattern will change.

Implementing NN's (I think) one may be more able to follow the pattern easier and faster - while the catagorizer actually may try decide which pattern is playing out at this moment.

Just my thoughts.

Hi. Just wondering since you have been using NN for trading this long, whether you are using commercial solutions such as TradingSolutions or Neuroshell?

From what you describe, probably not!

 
omelette:
Hi. Just wondering since you have been using NN for trading this long, whether you are using commercial solutions such as TradingSolutions or Neuroshell? From what you describe, probably not!

In the beginning, I also bought some of the available commericial software - don't we all??

But in fact - I am running my own software, my own structures etc.

Although I still buy "interesting" software if/when I think it's worthwhile.

For instance - my wavelet package (which I don't think I'd be able to make a success without) is a commercial addon. I also recently bought that SSA addon (which, to be honest, I don't use live - either the software is not as promised or I don't know enough about it yet.)

All I really wanted to say is - Neural nets are part of my software and they work for me.

Having said that - took me a lot of effort to understand how to implement them in this environment and so I thought that, perhaps I can save someone else a few years (and a lot of capital) by sharing some of my experiences.

 
omelette:
I also had a go (way back) at coding NN but after a few futile attempts at coding back-propagation, I decided to leave it to the experts!

Thanks for more info. Now I remember why I still not doing anything about NN. Probably after reading half a page of introduction to NN, then fell asleep . May be I had made a same conclusion as yours back then - Leave it to the experts!

Regards

Davod

 
davidke20:
Thanks for more info. Now I remember why I still not doing anything about NN. Probably after reading half a page of introduction to NN, then fell asleep . May be I had made a same conclusion as yours back then - Leave it to the experts!

Regards

Davod

Then again - we all know that in this environment:

there are no experts ....??

 
jdpnz:
Then again - we all know that in this environment: there are no experts ....??

No, I mean the frustration of trying to understand how the system works. In other word - LAZINESS But hey, who knows someday I'll pick it back up.

Regards

David

 
davidke20:
Thanks for more info. Now I remember why I still not doing anything about NN. Probably after reading half a page of introduction to NN, then fell asleep . May be I had made a same conclusion as yours back then - Leave it to the experts!

Regards

Davod

Hee hee, I think I have that book around somewhere in PDF form.

Take comfort from the fact that you have at least read half a page more of it than I have!

 

The working somewhat similar to Cyberia trader. Isnt it?

 

results

Hi,

No it's not based on cyberiatrder, I looked at that ea but it only gave good backtests not live.

Updated results:

Account 1: $13,994 (almost 40% increase) no loses

Account 2: $10,900 (9% overnight)

It looks good so far

 

attached the statement image...

Files:
 

Positive even today

Maybe you all have already seen. Today other positive trades closed by tp where "better's trades" failed. maybe it doesn't mean anything, but could be a good sign.

Metcalfe

Reason: