Machine learning in trading: theory, models, practice and algo-trading - page 1321
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
Clarify what? Yandex product yes.
Thank you. I was a little confused by CERN, I thought we were talking about a different product. I wondered if it was a different product.)
Well, and the phrase -"Alexei chose not the worst catbust, one of the best for research, by far.", implies that there are a lot of these catbusts.
In general, we have dealt with it. There is only one catbust.)
It's the clowning that pisses me off.
It's not what you think that's infuriating, it's that there are those who are so intent on meaningless kotir conversions
Yes, thinning will only lose some of the information about the price movement, no more
it's like looking at the price in 1972 and today and cutting out the restIs this about items other than the second one, or is it about my actual state from today?
Nice! When's the signal? Go to the branch TP and lead it, eh? I'm sick and tired, I have no results :((((
I'm pissed off by brainless people in a thread where they shouldn't be, nothing else.
What do you mean?
you mean those who do not rule with their own brains, but ask for help from neurons or what?
I'm out of my mind.
no supercomputer (or NS) will ever change a person's mind, 100%Hey, man! When's the signal? Go to the branch of Tip and lead it, eh? I'm sick and tired, and the results are still no :((((
I do not see the point in the signal.
Now the conclusions.
1. The distribution of kotir increments relative to the balance of supply and demand is of course and mirrored relative to the balance line...
2. The maximum increment does not change since the beginning of Forex (from 40 to 120 pips, depending on the pair)
3. The price has a memory, but the memory fades with time.
Is this about points other than the second or about my real state for today?
Rinat, is it NS? If it is, do you set stop-losses?
Rinat, is this an NS? If so, do you set stoploops?
No, not NS.
No stoplots and takeoffs.
I was just responding to a challenge thrown in the direction of all the supposedly stupid
As recently as yesterday we were talking about predicting sinusoids and I remembered my old topic:
Forum on Trading, Automated Trading Systems and Testing Trading Strategies
How do your TSs (forests and neural networks etc.) predict the sine waves?
Yuriy Asaulenko, 2018.05.22 16:39
A considerable part of the Machine Learning topic... is devoted to prediction-prediction. I propose a simple test to check the ability of your scaffolds and neural networks to predict.
So, you need to predict several samples (as far ahead as you can) of a simple analytic function on an infinite interval.
Y(t)=(1+0.5sin(t))sin(6t) (1)
The problem is not invented. It is a demo-example of one of MLP-type neural network packages. This problem is successfully solved by a simple NS.
The same package contains more complicated instances, such as extraction by a neural network of speech from noise, and also ordinary not very complicated MLP. You make the noise yourself, train it yourself, and isolate pure speech yourself. Impressive, in general.
Can your scaffolds and NS handle this task (1)? If not, what market forecasting can we talk about?
I am vindictive and I write everything down.)
I must say that the topic was in vain, but it got stalled and never reached the case. Perhaps because of the not very. clear wording.
Actually, we don't need to solve any problem. We take a function, such as the one in the thread, or even better a more complicated one. Create an artificial tool with this function and run it in the tester on an already working strategy. In theory, the profit should go off the charts on the working TS. Oh, I forgot, the function should be normalized in advance so that it approximately corresponds to the symbol to which the TS is tuned. Then we can add some noise and see what will happen.
I don't make forecasts and I don't have such TS ready, so I cannot check it in the nearest future. But in the distant future, I plan to.
Now about why we need all this.
Suppose we need to teach NS (or other MO) forecasting. Usually initial weights of NS are initialized randomly and if the NS gets into min-maxes when training, it's a big question.
Let's do the following.
1. We generate a non-random function close to market BP and use it for training of randomly initialized NS. Check it and so on. Now our NS is close to what we need in terms of settings, but so far it has not been able to solve the real problem.
2. Conduct training of the NS (see point 1) using real BP. At the same time we already have some guarantees that preliminary NS settings are already somewhere in the vicinity of min-max areas, and during additional training they will go where they should, but not to some random min-max.
The analogy is with a high school student who is first taught to solve simple problems on a topic, and then those problems are made more difficult. Teaching à la schoolboy is more effective than trying to make you solve complex problems at once.
In general, the method is not an opening, somewhere in the literature it was found, but there are many books, and I am alone - I do not remember. In any case, I thought about its implementation. Well, the very first experiment with the attempt of the ready TC to predict the analytical function, in general, is necessary, as staged.
No, not NS.
No stops or takeoffs.
I just responded to the challenge to all supposedly stupid.
Okay. Because I wanted to talk about supposedly "correct" trading style, we are talking about stops :).
Ok. Because I wanted to talk about the supposedly "right" style of trading, we are talking about stops :).
It's a very philosophical topic.
Stops are from the manual.
Signal change is by market