Trading Systems Leaders in this forum (automated trading systems) are winning more than 3000 pips in a month (30000$ investing one lot every time). Click here to register and get more information
this thread looks a bit dead, but I hope there is still people interested in Neural networks around! So please share your experience with NNets here!
I already tested a few combinations of factors to test the prediction power of neural nets.
As I read in different papers and posts, using previous High-Low-Open-Close seems totally unefficient. The NN is totally lost in the dark... The Fitting plot of real value versus NN output is horizontal (1st pic Open-Low-High-Close.jpg), meaning it can't predict anything at all. It is the same result as trying to find the loterie numbers with a NN
I tried different inputs and output values, normalized or not, the result is the same.
So what's worse to look at?? It seems that technical indicators work way better. I tested a mix of classic indicators as inputs to predict the absolute price. The fitting plot is not the best but not bad(2nd pic - Technical factors inputs.jpg). I also think using absolute values is a very bad idea.
Using the Moving average slope to predict future slope gives interesting results to identify possible reversals (3rd pic MA Slope.jpg).
Another interesting way of research is too use a NN to evaluate good trading conditions. So I created a normalized score claculated from the current close and it's distance to future High and Lows. If the score is high, it means that in the next 4 periods, the distance between current close and future highs is high (Buy profits aheads) and distance to low is low (no significant drawdown). I had no sucess with this method for the moment.
Hope to have some comments and experience sharings very soon!
What exactly were your inputs/outputs for the NN that utilized the moving averages?
Answers to OnTheRoad, aegis, MrM and anybody interested in NN's
Thanks OnTheRoad
Yes you are right, increasing the number of neurons can lead to better accuracy, but just slightly. If you have bad results with 20 hidden neurons, it won't be perfect forecast with 200 neurons for sure. You also have to be careful about not overfitting the NN : if you use a lot of neurons in the NN, you have to feed the NN with a lot of data. Matlab NN toolbox doesn't calculate an ajusted Rē to control over-fitting issues.
aegis : I used a few MA slopes (Ma value(t)-Ma value(t-1)) to predict the MA slope a few bars ahead. I'm not really sure about the trading power of this tool as I didn't backtested the system. But it looks it can take trades before the MA cross happen wich is slightly better compared to a basic MA cross system.
MrM: I didn't implement a dll solution. I coded some NN's under MQL4 as I found no real advantage from linking Matlab results with MQL4 as the NN doesn't need to be retrained every seconds. I just get the weights from matlab and copy-paste in MQL4
I was off for a few days, so I have no fancy graphics to post today!
I am looking for anyone who have a working Cortex Neural Network Software ( i think the last known version was version 5). If you have a working application please let me know. i need it very urgently to test and train some EA in metatrader 4 . I did purchase a copy from Neural Networks Tutorials and Software but it seems to be missing a "wait.dll" file. They also have a free copy but both of them don't seem to work since they are missing the "wait.dll"
Anyone who has used or have experience with this software . I am basically using it to train and test my Metatrader 4 EAs
I am running it on Windows XP professional with service pack 2.
Any help or suggestion is well appreciated. i need to Finish the EA by this weekend
Only just found your message.
I have been using Cortex for some years without problems. However - I don't have a "wait.dll" with the Cortex software and neither on my machine anywhere so, perhaps you have something else wrong??
I have been running Cortex on XP, win 2000 and years ago on 98 - all worked reasonably well for me so, I doubt if it's anything to do with your XP system...
My partner and I have created the dll between Matlab and Metaquotes...that was the easier part of the NN quest. It is finding the right net to implement. Maybe you can give us some feedback. We have tried classification, prediction, dynamic and probablistic networks. Lots of trial and error.
what I've learned so far is that econometrics are not usefull for short-term prediction (forget ARMA etc.), but Matlab is the way to go if you want to gain an edge, if you have all toolboxes and a good theoretical background it's a lot of fun.
Nonlinear methods (like NN, SOM's,...) are great, but don't try to predict anything: I think it's a bridge too far for today's technology, but prediction is not necessary: you are perfectly model the final bar of a timeseries in function of the past bars to see if you are close to a relative extreme value or not.
But the learning curve is not steep. I lost track of how much I spent on amazon for books filled with greek-letter-financial-math
Answers to OnTheRoad, aegis, MrM and anybody interested in NN's
Hello everyone,
Linear Mode
