Machine learning in trading: theory, models, practice and algo-trading - page 3364
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
It's much simpler than that.
Something was fitted to some part of a non-stationary random process without realising that any part of a non-stationary process has NOTHING to do with any other part of the non-stationary process. That is why the results at other segments are arbitrary: they may be good, but they may be bad, but in reality the sandwich ALWAYS falls down in butter.
By the way, the concept of "variance" refers to a stationary random process.
What does a non-stationary process have to do with looking for recurring inefficiencies? For example, the majority of scalpers-channelers have all sorts of scalpers who trade at certain times, where they are most predictable. And just at those times the process is quite stationary, otherwise a profitable TS is impossible. I am immune to such strategies because of constant wars with brokerage centres, but nevertheless they exist. It is just a kind of a game on the peculiarities of quoting, basically. There may not be much need for MO there either.
I am discussing the posted chart and commentary on them, including yours, not your scalpers in your pocket.
I am discussing the posted chart and commentary on them, including yours, not your figure in your pocket in the form of scalpers.
It's much simpler than that.
I'm relieved.
Maybe it makes sense to look for a parameter and a formula by which to calculate your optimised parameters. Based on the results of the optimisation. Of course it's complicated.
I don't get it, unfortunately.
In the last article, he suggested a variant on how to make training more stable for MOs. That is, there is less retraining. But profitability suffers.
Some also use the term "degree of freedom". The higher the degree, the higher the probability of fitting. And it grows non-linearly.
Didn't realise anything, unfortunately.
Well, it is when the stoploss is calculated as yesterday's atr. Dynamic means depending on some parameters. In the terminology of mytarmailS
This is great when the return on Sample is a lot less, but stable on OOS.
Some still apply the term "degree of freedom". The higher the degree, the higher the probability of fitting. And it grows non-linearly.
It is not always linear, there are also steps of quantity to quality.
Well, it is when the stoploss is calculated as yesterday's atr. Dynamic means depending on some parameters. In the terminology of mytarmailS
Then according to such terminology the ATR parameter will be a constant. I do not understand such a view of optimised parameters.
Then according to this terminology the ATR parameter will be a constant. I do not understand such a view of optimised parameters.
The idea is to find more stable for profit parameters, which regulate the optimised parameters. In addition, there is a possible feedback of influence on the parameters. But this is not an end in itself, but as a branch of research.