Discussion of article "Third Generation Neural Networks: Deep Networks" - page 16
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
Hi Vladimir, I am very impressed with your article.
I managed to get it installed and tried with the various steps in R. I have some doubts about spatialSign transformation hopefully you can help me to understand.
I tried to learn the effect of preProcess with spatialSign, so I tried the following codes:
predict(preprocess(try,method="spatialSign"),try)
I get the following results:
[1,] -0.7071068 -0.7071068
[2,] -0.7071068 -0.7071068
[3,] 0.7071068 0.7071068
[4,] 0.7071068 0.7071068
I was very surprised with this result, intuitively, I would expect 1 and 2 should not be the same in spatialSign. I know it first center and scale then apply spatialSign, is the result correct?
Vladimir, thank you for very interesting material. A year ago I wrote a model of the system using FireBird database as an intermediate storage of data and signals between R and Expert Advisor, but the system was used not on MT5, but it doesn't matter. It gave me the opportunity to experiment with other R algorithms. Thanks again.
Hello , I am from China , I am always very concerned you in mql5.com articles . The four articles are very research for me , the cost of learning . I admire your professional knowledge . Thank you as a reader can share your experience . in your article there is always a place where I am confused , please give me an answer at your spare time . Thanks !
why pr.sae>mean(pr.sae) Yes sig=-1 not sig=1 ?
Hello , I am from China , I am always very concerned you in mql5.com articles . The four articles are very research for me , the cost of learning . I admire your professional knowledge . Thank you as a reader can share your experience . in your article there is always a place where I am confused , please give me an answer at your spare time . Thank you !
why pr.sae>mean(pr.sae) Yes sig=-1 not sig=1 ?
When we defined the target variable we assumed that 0 is BUY , 1 is SELL/ ?
# ЗигЗаг имеет значения (определен) на каждом баре а не только в вершинах
zz<-ZigZag(price[ ,'Med'], change = ch, percent = F, retrace = F, lastExtreme = T);
n<-1:length(zz);
# На последних барах неопределенные значения заменим на последние известные
for(i in n) { if(is.na(zz[i])) zz[i] = zz[i-1];}
#Определим скорость изменения ЗигЗага и сдвинем на один бар в будущее
dz<-c(diff(zz), NA);
#Если скорость >0 - сигнал = 0(Buy), если <0, сигнал = 1 (Sell) иначе NA
sig<-ifelse(dz>0, 0, ifelse(dz<0, 1, NA));
return(sig);
}
fantastic article indeed even in these days.
but I got a question why my Kzz equals -Inf？
sig.zz<-ifelse(tail(dt[ , ncol(dt)], 500) == 0, 1, -1)
bal.zz<-cumsum(tail(price[ , 'CO'], 500) * sig.zz)
Kzz<-mean(bal.zz / bal)
Kzz -Inf
Using Rsi could be the wrong approach.
Maybe its a better one to let it trade directly, like as if it needs to learn to walk, or play chess or any other game.
pr.sae<-nn.predict(SAE, x.ts)
sig<-ifelse(pr.sae>mean(pr.sae), -1, 1)
sig.zz<-ifelse(y.ts == 0, 1,-1 )
bal<-cumsum(tail(price[ ,'CO'], bar) * sig)
bal.zz<-cumsum(tail(price[ ,'CO'], bar) * sig.zz)
sir, above code, when calc bal you havent move the sign backword like you did in the ariticle DEEP NEURAL NETDEEP NEURAL NETWORK WITH STACKED RBM.
sth do I miss?
NEW ARTICLE The third generation of neural networks: deep networks have been released:
Author: Vladimir Perervenko
the target is worng ,it cannot work in realtime !!!!
the target is worng ,it cannot work in realtime !!!!
Don't write stupid things. Please read this article carefully. Think about it. After this, a repetitive example that proves its point.
Throwing empty slogans at many people.