Machine learning in trading: theory, models, practice and algo-trading - page 3459

New comment
 
fxsaber #:

Forced to come up with stationary transformations with loss of information because you know that MO does better with stationary data.

So you get the chicken and egg paradox. You lose information on purpose to make it convenient for MO. And the output is nil. Or you submit data that is inconvenient for the MoD and you get zero again.

 
Any TC requires stationarity for a stable result.
 
Maxim Dmitrievsky #:
Any TS requires stationarity for a stable result.

Perhaps somewhere stationarity manifests itself on profitable TS not in an explicit form - stationarity is not created on purpose.


For example, someone trades only in the evening session because it is profitable. He has no idea about the reasons of this fact. It is just a fact.

If we start digging into the reasons, it may turn out that the market is more likely to be stationary in the evenings than at other times.


But the man who discovered evening trading didn't even know anything about the concept of stationarity. He never looked for it. He just tried it and it worked. This man is much dumber than MO. But for some reason he had a primitive idea to put a trading time limit on his stupid TS.

 
fxsaber #:

Perhaps somewhere stationarity manifests itself on profitable TCs not in an explicit form - stationarity is not created on purpose.


For example, someone trades only in the evening session because it is profitable. He has no idea about the reasons of this fact. It is just a fact.

If you start digging into the reasons, it may turn out that the market is more likely to be stationary in the evenings than at other times.


But the person who discovered evening trading didn't even know anything about the concept of stationarity. He never looked for it. He just tried it and it worked. This man is much dumber than the MoD. But for some reason he had a primitive idea to put a trading time limit on his stupid TS.

Well, people like that happen
 
Maxim Dmitrievsky #:
Well, there are people like that

There's no such thing as an M.O.

 
fxsaber #:

There is no such thing as an MOE.

I lost the thread of the narrative :)
 
Yeah, this conversation's a little weird.
 
A time filter is almost equal to a volatility filter with a period of 20. In the last article there is an example of such a TS on MO.

Even in earlier articles there were time filters.

I gave up time because there is clustering of volatility in time.

I did not see any other hints.

 
Maxim Dmitrievsky #:
I didn't see the other half-hints.

Whatever.

 
fxsaber #:

Whatever.

🙀
1...345234533454345534563457345834593460346134623463346434653466...3699
New comment