Machine learning in trading: theory, models, practice and algo-trading - page 278
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
Of course, I accidentally found a deal in my diary from 5 years ago, the time when I was trading on "oanda". It was a good time, full of hope and some kind of market romanticism. :)
If you have an understanding of what is savings / allocation, understanding who you take the money from, and most importantly discipline and patience, not only to enter successfully, but also to wait out the entire movement, with the last points I have chronic problems, and that's why I joined the algorithms :)
P.S. the whole trading system in one post FREE :))
No, it's not a mistake.
rattle is a pretty mean-spirited joke.
On the one hand it allows you to see the whole cycle of machine learning: dataminig, modeling, model evaluation at a fraction of the cost.
On the other hand, it gives the illusion that models are easy to use. And this is not easy at all. Each element of the models used requires a very thoughtful attitude. And this is required at absolutely all stages of work on MI application. In particular, the choice of a target variable is not a simple problem either. It's possible to choose very tempting target variables and then it turns out that it's impossible to select predictors for them. If we talk about the ZZ in the article, strictly speaking, it is not clear at all WHAT is being predicted....
Success, and most importantly without illusions, there is no grail, you need work and knowledge.
The fact that you have already taken the leading indicator, it has already shifted into the future. If you took the Momentum indicator, it would have to shift by half a period, but ZZ shows the future at a particular time. To trade with ZZ it is not necessary to shift it, if it is not repricing? Window statistics with a lag should be shifted, ZZ has no lag, the slope of ZZ shows the ideal position direction and should be taught to the classifier.
The shift will not significantly spoil the learning, if the distance between the knees ZZ is much larger than the step of the shift, but you will teach the system ahead of ZZ, which means to teach to work "better than perfect", this will worsen the learning, for obvious reasons, because the relevant signs may not have been formed yet.
For pruf I advise to take any simple series, a sinusoid for example, with one trait for example momentum in half period of sinusoid, targetZZ with and without shift and teach the forest, you see that with a shift will be a little bit worse than without.
The point is that you have already taken a leading indicator, it is already shifted into the future, if you took momentum, then yes, you had to shift by half a period, but ZZ shows the future at this particular moment. To trade with ZZ it is not necessary to shift it, if it is not repricing? Window statistics with a lag should be shifted, ZZ has no lag, the slope of ZZ shows the ideal position direction and should be taught to the classifier.
The shift will not significantly spoil the learning, if the distance between the knees ZZ is much larger than the step of the shift, but you will teach the system to be ahead of ZZ, which means to teach to work "better than perfect", this will worsen the learning, for obvious reasons, because the relevant signs may not have been formed yet.
I suggest to take some simple series, a sine wave for example, with one trait for example momentum in half period of sine wave,ZZ target with and without shift and teach the forest, you will see that with shift will be a little bit worse than without.
It would be desirable to see results of calculations
I hope you have read all the articles that have already been published on the site on these topics and will not describe the invention of the bicycle?
New ideas are always interesting.
Good luck
Yes even a monkey for three rubles can do it here, this is a typical descending flag.
AHAHAHAHAH.... Yeah )) and there was a typical stochastic in a typical overbought zone and most likely a typical trend line that was typically broken and something else typical that I missed... And there was also a typical analyst who typically analyzes price after the fact..... Typical victim of another forex kitchen...
ahahaha.... Yeah )) and there was a typical stochastic in a typical overbought zone and most likely a typical trend line that was typically broken and something else typical that I missed... And there was also a typical analyst who typically analyzes price after the fact..... Typical victim of another forex kitchen...
Typical downtrending flag, purchases were closed new sales were opened what's the big deal. Actually there are 2 flags there. Typical directional sales, typical directional pullback, nothing new in the market and won't be.
Typical downtrending flag, purchases were closed, new sales were opened, what's the big deal. Actually there are 2 flags. Typical directed sales, typical directed pullback, there is nothing new in the market and will not be.
Once again, everything is typical after the fact, after the fact you can explain the market behavior by a stochastic, by a news and figure or any other bullshit, you can even explain the fall by your left ear itching and if it sounds plausible to hamsters, they will believe it and this fable will work, but you know what? Tomorrow, when the signal will repeat and the decision will have to be taken real time instead of post factum. All hamsters will lose money on this typical signal and this situation will be typical for the market.
I can find millions of examples where your flag didn't work, so what's the typicality here? What's the predictive power? Moreover, I can generate a random noise and give it to you, saying that this is the eurodollar and you will find hundreds of your descending flags there, does that mean that they are there? you don't even realize it's not eurodollar, think about it.
I will repeat it again, post factum is typical, post factum you can explain the movement by a stochastic, by a news and figure or any other bullshit, you can even explain the fall by your left ear itching and if it sounds plausible to the hamsters they will believe it and this fable will work, but you know what? Tomorrow, when the signal will repeat and the decision will have to be taken real time instead of post factum. All hamsters will lose money on this typical signal and this situation will be typical for the market.
I can find millions of examples where your flag didn't work, so what's the typicality here? What's the predictive power? Moreover, I can generate a random noise and give it to you, saying that this is the eurodollar and you will find hundreds of your descending flags there, does that mean that they are there? you don't even realize it's not eurodollar, think about it.
Do you see the directional movement? You see, so according to you participants relied on your conclusions at that moment. (All the moment they all started to believe the main use of the scientific work of forum member mql5 (mytarmalis)) Do you believe it yourself?
Well, let's have a laugh!!! ))) Really, each about his own, well, that's the thing....)
Well, let's have a laugh!!! ))) really each to his own, well that's it....)