From theory to practice - page 966

 
Alexander_K:

In temporary windows. You have some kind of windowless TC at all - am I right in assuming that?

That's right, it's fully adaptive to any tool.

It's kind of like machine vision but much simpler.

Works everywhere. Always right.

I already checked. The main thing is there are no parameters. The market draws my system by itself. As I said before, the market evaluates it relative to the market itself.

The problem of losing orders, however small they may be, has not gone away as it should.)

 
Martin Cheguevara:

But you wrote that you use multiparametric statistical methods of calculation...

Physicist?

 
Алексей Тарабанов:

Physicist?

a little bit of everything...)

Programmer most of all. Although physics, statistics and mathematical analysis have always been my favourite subjects.
 
Martin Cheguevara:

But the problem of losing orders, no matter how few there are, has not gone away).

So here comes to the aid of Asaulenko, the most boring radio amateur, who says: Well, you made a loss, cut it, so what now... Here I agree with him. Chasing a break-even profit leads to disasters - it has been proven on the Automat tractors. Don't repeat his mistakes...

 

Summary: A correct, measured entry into a trade is far more important than an exit from it.

Amen.

 
Alexander_K:

So that's where Asaulenko, the most boring amateur radio operator, comes to the rescue, saying: well, you made a loss - cut it, so what now... Here I agree with him. Chasing a break-even profit leads to disasters - it has been proven on the Automat tractors. Don't make the same mistakes...


you see any excesses here?

You have to close on them anyway. It's a rough signal that it's going to go either way.)

 
Martin Cheguevara:


do you see any excesses here?

It's a rough signal that it's going to go either way).

If you trade on the trend - yes. But still, there are differences there. In 2 cases the kurtosis is <100 (conditional) and in the two most extreme ones >>100.

 
Alexander_K:

If you trade on a trend, yes. But, still, there are differences there. In 2 cases the excess is <100 (notional) and in the two most extreme cases >>100.

Yes, but as you see in the two >>100% one signal went to the rebound and the other went down, so you have to close in any case

 
Martin Cheguevara:


In short, I understand that you have a TS that gives 100% p.a., but that's not enough for you, right? And you are trying to solve this problem by moving to a full break-even point?

Most dangerous way! (see post above). Maybe, just trade on a larger number of currency pairs?

 
Alexander_K:

In short, I understand that you have a TS that gives 100% p.a., but that's not enough for you, right? And you are trying to solve this problem by moving to a full break-even point?

Most dangerous way! (see post above). Maybe just trade more currency pairs?

I'm trying to solve the fundamental problem of escorting losing orders. In itself, this function is very important to all of us, regardless of the number of profitable trades and % p.a.

If it is solved, you can deposit a thousand dollars in your account and sleep well at night)

Reason: