Machine learning in trading: theory, models, practice and algo-trading - page 1444
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
I tried to study the meaning of the statement about the disappearance of series memory when going to differences. There is a reference to the book of the auntie professor, where I have not yet seen such a statement.
I tried to study the meaning of the statement about the disappearance of series memory when you go to differences. There is a reference to the book of the aunt professor, where I have not yet seen such a statement.
This is very simple, because memory is a sequence of something and not the difference of 2 nearest values
https://ru.wikipedia.org/wiki/%D0%9F%D0%B0%D0%BC%D1%8F%D1%82%D1%8C
paradoxically, even such a seemingly simple concept is sometimes better studied in more detail
because everyone intuitively takes it in some very simple sense
this is very simple, because memory is a sequence of something, not the difference of the 2 nearest values
You Teachers at the second stage are so funny))) hilarious...
This is very simple, because memory is a sequence of something, not the difference of 2 nearest values.
Frankly speaking, I don't really like the term "memory" in this case - too many unnecessary associations. Essentially, we are talking about the presence of stochastic dependence between the members of a random process, one of the realizations of which is our particular series. This dependence is defined through all possible joint distributions of members of a random process - in fact, all these distributions (by definition) define a particular random process. In reality, however, we are given a single realization of a process (our price series), which alone does not provide sufficient information to reconstruct all those distributions. That is why we have to additionally apply restrictions to the process describing our prices.
In general, the transition to differences is a very special case of linear transformation and will allow to get rid, at most, of linear dependence. If we could get rid of dependencies so easily, we wouldn't need to bother with non-linear PCA, for example.
You Teachers in the second stage are so funny))) hilarious...
Although you are far from being a role model, but I have to agree with you at this time, denisenko is going the way of v-suspending to workers and teachers of 100-500$ plumers, this is mean, I would even say that this is a natural cattle and he does not care about the net, he probably will never show it again, because it is clear what it is, but will tweet about the zero error, which means nothing.
To be honest, I don't really like the term "memory" in this case - too many unnecessary associations. In fact, we are talking about the presence of stochastic dependence between the members of a random process, one of the realizations of which is our particular series. This dependence is defined through all possible joint distributions of members of a random process - in fact, all these distributions (by definition) define a particular random process. In reality, however, we are given a single realization of a process (our price series), which alone does not provide sufficient information to reconstruct all those distributions. That is why we have to additionally apply restrictions to the process describing our prices.
In general, transition to differences is a very special case of linear transformation and will allow to get rid, at most, of linear dependence. If we could get rid of dependencies so easily, we wouldn't need to fuss with non-linear PCA, for example.
mbm, later I'll finish and compare the results, without making up anything unnecessary
I bought neurosoft for forex trading. After training gives out such a picture from the beginning of 2018, where for training took 2018, and the last section (from January 1 to date - three and a half months) - is a forward.
What I want to say: this thing works, it makes sense to dig in (all you neural network experts). I will test it in real life.
I bought neurosoft for forex trading. I bought neurosoft so I can use it for trading forex. After I finished training it shows the following picture: from the beginning of 2018, 2018 was taken for training, and the last position (from January 1 to this day - three and a half months) is a forward one.
What I want to say: this thing works, it makes sense to dig in (all you neural network experts). I will test it in real life.
What kind of software is it, how much did it cost?
And what kind of software was it, what did it cost?
Some team of mathematicians at the institute, decided to try to exploit their developments in the financial markets. I bought it for a lot of money, I will not reveal the details. I am not going to buy some derivatives but I am sure I can buy them with my own money.