My EA's test results - what's the catch and illusion? - page 7

New comment
 
avtomat:


No!!!

I suspect you are only looking at flip strategies, ie OpenBuy --> CloseBuy=OpenSell --> CloseSell= OpenBuy --> etc.

But this is fundamentally wrong. Because both signals CloseBuy != OpenSell and CloseSell != OpenBuy --- these signals are not equivalent (flip equivalence of signals is not a rule, but an exception to the rule)

The system must consist of two closed subsystems OpenBuy --> CloseBuy & OpenSell --> CloseSell

Let's focus on CloseBuy and CloseSell signals. Suppose they are not equal to OpenSell and OpenBuy signals. Then they are independent and do not correlate in any way with the opposite entry signals. But since they are signals, they are significant and their MO is positive, so they can be used as a separate system that will enter the market by these signals as if hedging the previous system. If we start claiming that these signals are valid only if they are closing signals, i.e. if the opposite buy and sell trade was executed before, then we get complete nonsense because the market does not know whether we have any position at the moment or not. I continue to argue that there is no connection between entry and exit signals, hence no criteria for an exit TS, but a second TS hedging the first.
 
TheXpert:
So you probably don't understand the concept of "signal"


Rather, it means I don't understand the meaning YOU give to the concept of "signal".
 
C-4:
Let us focus on CloseBuy and CloseSell signals. Suppose they are not equal to OpenSell and OpenBuy signals. Then they are independent in themselves and do not correlate in any way with the opposite entry signals. But since they are signals, they are significant and their MO is positive, so they can be used as a separate system that will enter the market by these signals as if hedging the previous system. If we start claiming that these signals are only valid if they are closing signals, i.e. if there has been a contrary buy and sell trade before, then we get complete nonsense, because the market does not know whether we have any position at the moment or not. I continue to argue that there is no connection between entry and exit signals, hence no criteria for an exit TS, but a second TS hedging the first.


What has MO, correlation and other statistical swear words got to do with it ;))) And what the market knows or does not know is also irrelevant. And hedging has nothing to do with it.

The point is that a signal to close a straight position is NOT equivalent to a signal to open an opposite position. And vice versa. These signals have different functions.

 
This is probably a case where one needs to know the essence of the system in order to discuss specifics... In my opinion, it would be logical to "close on the rebound" regardless of further consideration of hairpin, flat, or"continuation of movement" options. "Close" = "open" - there is a bid for "maximum efficiency" which is not achievable "by the original rules".
 
avtomat:


What has MO, correlation and other statistical swear words got to do with it ;))) It has nothing to do with what the market knows or does not know. And hedging has absolutely nothing to do with it.

The point is that a signal to close a straight position is NOT equivalent to a signal to open an opposite position. And vice versa. These signals have different functions.

That's what I'm talking about. The functions are different, which means different TS, full stop.
 
avtomat:

...

The point is that a signal to close a straight position is NOT equivalent to a signal to open an opposite position. And vice versa. These signals have different functions.


Exactly right. Opening increases the risk, while closing reduces it. Consequently, the requirements for an opening signal (flip) are much greater.
 
C-4:
That's what I'm talking about. The functions are different - hence different TSs, full stop.

I completely agree. I also came up with the idea that any complex TS can (and should) be broken down into separate components, the simplest TS. And study the statistics for these simplest TS, select the optimal parameters, etc. And only after that make a final product out of these bricks.

The point is that a signal for closing a straight position is NOT equivalent to a signal for opening an opposite position

Imho, it's simple. Each signal at any point has some force (statistical advantage). When this force is equal to a certain value, we open a position. If this value becomes zero (no statistical advantage), we close the position.
 
Meat:

...

Imho, it's simple here....
Only cats are born simply...)
 
The topmistress seems to have been blown away... He's probably making millions by now-)-)-)
 
khorosh:
It's just that cats are born... )

It's especially hilarious to read this from a martinist )

Meat:
Imho, it's simple here. Each signal at any point has some strength (statistical advantage). When this strength is equal to a certain value, we open a position. If this value becomes zero (no statistical advantage), we close the position.
Uh-huh, the whole point is in two lines.
12345678
New comment
Reason: