Machine learning in trading: theory, models, practice and algo-trading - page 1793
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
In fact, the usual matstat is a statistical hypothesis test. For example, let us have only one of two possible models - SB or Ornstein-Uhlenbeck, then we get the problem of distinguishing the two hypotheses, which is solved by the well-known Dickey-Fuller test.
Greetings Colleagues,
I understand that the question is completely off-topic, but still I will ask. Does anyone have an image of Ubuntu to upload it to OpenStack? Well, that the image was configured and started to boot on a remote server. Ubuntu need a desktop, of course, a graphical mode, and then I again have fried, frankly. 6 hours loaded into myle image of his system, but ustrazhaetsya on the hosting he did not, when he started to understand that it had to configure the tricky way. Now I made another image of the type of settings, but I don't know if it will work or not. And the main thing I ask the technical support of Maylova "make at least one standard image with graphical desktop, it would be just right for machine learning", and they answered me: "We do not configure operating systems of our clients". What's the use? I spent all my money on it, but I still couldn't install the Instant in graphical mode. It's just a mess, comrades :-(
Well, what do you think? It means I got stuck in the printer, whether it was jam or a cherry plum for 15 minutes, and I had no idea. And I had a chance to pop up on the forum that would be 5 seconds to be thrown out and not available for several hours. It happens :-)
Try pouring vinegar on it, see if it sticks.
Try pouring vinegar on it to see if it sticks.
You're funny :-) I was planning to wash my keyboard in a washing machine, but I like your way better :-)
then set it to 90 degrees and more bleach
The concept of root (characteristic polynomial) is defined ONLY for autoregressive processes. There are reasons to consider any stationary process as autoregressive. There are also nonstationary autoregressive processes. But there are many more processes which are NOT stationary and are NOT autoregressive (and not reducible to them in any way) - for them reasoning about roots makes no sense at all.
This is a necessary condition (but not sufficient) and it works only within the given two-state assumption. If it is not fulfilled, then it makes no sense to argue that we are dealing with a series different from SB (the introduction of the second state turned out to be redundant - the price is always similar to SB). If it is satisfied, then we need to check the normality and independence of the residuals, the significance of the difference between the parameters and zero, etc.
Well, yes, starting with their minimum and gradually building up, realizing that at some point everything will be perfectly "described" due to overfitting from the abundance of parameters.
The point is probably the primary definition of the SB series. And only then move on to various decompositions and searches.
The question is the minimum reliable plot for determining the SB. And what reliable and not very expensive way to do it.
The question of scale. In fact, the presence of regularities on a certain scale / TF indicates the presence of sufficient connection of external factors with the price, or sufficient influence on the price, at which these regularities could be detected. At that the period of influence of external factors on the price is commensurable with our scale. And with a sufficiently even mixing of factors and the degree of influence on the price, it is indeed acceptable to have situations where the influence cannot be separated and for us the behavior of the price will be SB. But judging by the behavior of prices, there are not many such states.))
The presence of autoregression, when the price depends on the previous value, what does it mean mathematically? The presence of dependence on factors, functional dependence, logical dependence does not mean the dependence on the previous values.
Take the test to distinguish the real from the neural network (fictional network)https://proglib.io/tests/pravda-ili-lozh-chto-umeyut-neyroseti
My score was 6 out of 10. I got a puncture on the pictures of people.
Take the test to distinguish the real from the neural network (fictional network)https://proglib.io/tests/pravda-ili-lozh-chto-umeyut-neyroseti
My score was 6 out of 10. I got a puncture on the pictures of people.
I got a more modest 2/10.
I have a more modest 2/10
I'm sure the result will be 50/50 if you take a large enough sample of test subjects