Where is the line between fitting and actual patterns? - page 9
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
The question is not about the amount of TO, but about choosing the size of the optimization period and the TO period so that those regularities that we find on these periods would work in the future. What are your thoughts on this?
And who knows. There are patterns that have worked throughout the history of the euro, for example. And there are laws that began to work at the beginning of the crisis. And there are laws that, on the contrary, did not work during the crisis, and then started to work again.
It does not depend much on the size of the period.
"pattern", the third most popular word, after grail... who can give an example of a pattern that worked and stopped.
This is not a question of the size of the OOS, but of choosing the size of the optimisation period and the OOS period so that the patterns that we find in these periods will continue to work in the future. What are your thoughts on this?
Before giving your thoughts, I would like to know what regularities you have in mind?
Regularities of the found sets of optimisation parameters? Or?
It's not a question of the amount of OOS, but of choosing such a size of optimization and OOS periods that the patterns found on these periods would continue working in the future. What are your thoughts on this?
It's the same thing, but from the side.
Representativeness - it has to be ensured. There should be 100-200 TC decisions on OOS. This may be a year on OOS and two days, it makes no difference, the main thing is reliability.
Then you have to choose the optimization period which is twice as long. You may have to (and often do so) vary the optimization period size.
Difficulties usually arise with TSs, whose frequency of trades and profitability directly depend on the volatility, because OOS period will also depend on the volatility.
I design TS which make decisions with frequency stable, which some metronomes would envy. That's why there are no problems with the number of deals in the optimized area and OOS, besides, calculation of statistical indices of the TS is much easier, because the number of deals is a constant.
Yes please. Some time ago on Wednesdays there was a local trend reversal in the Euro.
Thank you. Could you confirm this statement with figures and a table?
It's like this: ........ After the rain on Thursday it turns out..... ))
"pattern", the third most popular word, after grail... who can give an example of a pattern that worked and stopped working.
Yes please. Some time ago on Wednesdays there was a local trend reversal in the Euro.
good joke... i.e. a pattern. i too have watched the gaps overlap at the opening of the week.
who collects patterns? give me examples, so to speak, of disappearing patterns.
Thank you. Would you be able to confirm this statement with figures and a table?
It's like: ........ After the rain on thursday it's..... ))
And you check for yourself, download the day's candles and make a spreadsheet. Can you do it?
Of course I can and I always take responsibility for my statements if I dare to assert something.
For example today and here
........................................
Can you prove your claims?