OPTIMISATION as the basis of a trading system. - page 3
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
there's fuzzy logic implicitly and everything else.
)))) What does "stationarity" and "you can find patterns" have to do with it? There are patterns on non-stationary series as well.
"Any segment of history is essentially stationary" - what is "essentially"? I only knew before - stauionary and non-stationary series, but "essentially" - don't know.
It is not worth asking him such questions in order to protect the thread from more nonsense.
The author probably meant that you can try to guess the parameters of the TS, not the price.
The price, with all its noises and noise, is not what we need, but the profit. And any system can become profitable by varying its parameters in the necessary area.
then optimisation should be carried out on all timeframes, with the smallest possible step.
Check out this thread, preferably the last 10-15 pages.
I think it is unlikely that any CPU can provide such acceleration. Of course, not on every algorithm, so don't get too excited.
)))) What does "stationarity" and "you can find patterns" have to do with it? There are patterns on non-stationary series too.
"Any segment of history is essentially stationary" - what is "essentially"? I only knew before - stauionary and non-stationary series, but "essentially" - don't know.
In essence - it means that on historical data of any (I emphasize - any) length you can always find those patterns that will work over the entire length (I emphasize - on the entire) of the historical data. To work means to make a profit ))). Thus, there may be an impression that the market does not change, i.e., it is stationary. In this regard, there may also be an impression that these found patterns will continue to work. However, further on, these found patterns may stop working in the future. Immediately or not immediately is another question )))). That is the trick of financial markets.
Your rubbish, a priori of course, adds more clarity than mine ))))
Folks, I understand that stationarity has become a buzzword thanks to me and faa, but don't exaggerate - stationarity is only needed to determine the correctness of specific forecasting methods. It's just that most of mathematical statistics methods are developed exactly for stationary series. There are methods for non-stationary ones, but their number is very small.
There are regularities in the behaviour of non-stationary series and nothing prevents you from making money on them.