Using neural networks in trading - page 3
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
curious as to what people are going to input to get a prediction of the colour of the candle's body?
Wouldn't this do the trick?
My outline from 2010 seems to be from Ezhov, Shumsky.
P.S. I'm no expert in nets. :-)
Materials, so far, postponed...
That's it, if anything is unclear, please ask. Lines drawn manually based on the bottom indicator, assumed open
)))) It's just to inject some heat into the subject. See if this topic overtakes the euroflood )))) I'm sick of watching topics die before they are born.
:-)
It is useless! Here, I do not rule out that it will be more serious! After all, the NETWORK is NEURO!!! :-)
Von and Grasshopper brothers at one time in the beginning - they got off to a very good start and discussed the peculiarities of learning how to earn money ... :-)
Well, for example, in my screenshot there are red underlines. I don't see how it can be coded and networked at all.
The formulas with the squiggles on top are similar (to me) to what Yusuf had in his (18) in his article on exp...
These "second-order diffusers" :-) in terms of translation into code - confuse me... :-)
He's had an indicator made and an exp on his article...
Here, IMHO, there is also much to dig into... in terms of preparing data to enter the network... and not only...
What's the big deal? ))))) So far, one thing is not serious. They just put up pretty pictures, nothing more.
It's a big topic. Apparently no one is interested. But to dig deeper, there may be some people who answer.
Now we can Skype and discuss problems constructively and without unnecessary bickering
Now we can Skype and discuss problems constructively without bickering
You can do it here too, what secrets are there? For example, do you use third-party libraries or use MQL to teach networks?
MQL has too much complexity
and there is no debugging
MQL has too much complexity
and there is no debugging
I did not find any problems. The difficulties were with debugging, but otherwise everything is fine. Linear network with 8 hidden layers works fine, so far only on one pair. I can't beat the division to zero and I'm moving the whole network to all 28 pairs.