A question about making money in the FOREX market - page 12

 
C-4:
I am not comparing markets to casinos. I am not claiming that markets=casinos, in fact, I know they are not. But if they are not, then we need methods capable of proving it, identifying the difference between the two and building a profit-generating trading model based on that difference. If the supposed profit making method cannot even distinguish casinos from non-casinos, how is it going to make a profit? Where is the guarantee that he won't mistakenly sit down at the roulette table instead of at the money printing press?
Why are we comparing it? Why prove it? There's no need to bring casinos into it.
 
avtomat:
Why should we compare it? Why prove it? There's no need to bring a casino into it.

No, so there isn't. Let's assume that the signals of all indicators are unique, fluffy and accurate and correspond only to the market structure. The divergence appears - we buy or sell, it is the "whisper" of the market, it is a unique signal, it brings unique information, and everything else is inept.
 
C-4:

No so no. Let us assume that the signals of all indicators are unique, fluffy, accurate and correspond only to the market structure. The divergence appears - we buy or sell, it is the "whisper" of the market, it is a unique signal, it carries unique information, while all the rest is evil.

Well, what's the point of being sarcastic...? And besides, it's not my claims about "unique, fluffy, accurate, etc. etc." ;)

But then again, what's the casino got to do with it? ;)))

 
C-4:

What I like about this is that we have a model whose properties are only known theoretically and an unknown market, a small fraction of whose properties the model is supposed to exploit. Whether these properties exist or not is not known. The nature of market prices is also not entirely clear. There are too many uncertainties. Let us replace the SB market - now all its characteristics are known to us in advance. Let us test the model on it - yes, the model does not behave the way it is required by the SB - and there is only one explanation, it is the model itself, the inaccuracy or error can only be there and nowhere else. We correct it. We see the result. It corresponds to the requirements of the SB - good, the model corresponds at least to itself. Run the model on real rows. Compare results: there is a difference that must be corrected. Then we analyze the difference, and see how we can make profit and improve the model in the direction of increasing the difference between the IBB and the real series.
This is the second valuable idea I've picked up from you. We should add to it the reaction to the spike of two types: return and continuation of the spike (step).
 
Mathemat:
It is not uncommon in my systems to feed a SB with a Gaussian distribution of returns to see if they see fish in the SB. If some do, the system is dead.
That's my point exactly. When we use the system, we trust it. It sees something we don't. In the real world, we don't know what the market really is, and the only thing we have is faith that the system knows what it's doing. If it doesn't know what even we know, why use such a system? Better to trade with our hands, because we know more than it does and we can handle it better.
 
avtomat:

Well, why be sarcastic...? And besides, it's not my claims about "unique, fluffy, accurate, etc. etc." ;)

But then again, what does the casino have to do with it? ;)))


So you deny any at least outward "similarity" between the ball wandering charts and the price chart?
 

Reading - I can't understand anything.... It's too complicated...

You went to a casino and recorded the results of 1,000 ball rolls - that's "prepared the data"?? That doesn't make any sense. You didn't prepare it in any way. You analyzed it and noticed that zeros came out more often? So bet on zero! That's the prediction. On the other table, none of the numbers come out more often? Well, take it as a realization of even distribution and you're welcome to predict!

No one prepares any data, the prediction is in principle possible in both cases!

"The nature of market prices is also not entirely clear" - who is not clear? Don't you know the terms supply and demand? Don't you know the participants in the markets?

 
C-4:
That's my point exactly. When we use the system, we trust it. It sees something we don't see. In the real world, we don't know what the market really corresponds to.

I would like to develop this thought.

I refer to the basic requirements for the model as reversibility (not mine, according to Box). Let's take kotir = trend + noise. By level adding the right part we obtain the left one. It would seem that the reversibility requirement is satisfied. But it is not. Except for the kotir level there is something that we cannot see or model. The trouble is that we don't know if this "something" is lost in the right part or not. Could this "something" be what generates the non-stationarity? How do we find the black cat in a dark room, or is it not there?

 
Mathemat:

It is not uncommon in my systems to feed the SB with a Gaussian distribution of returns to see if they see fish in the SB. If any of them do, the system is dead.

There's no system like a system...

An SB with any distribution, including a Gaussian distribution, does not impose any bans.

C-4:
That's my point exactly. When we use the system, we trust it. It sees something that we don't see. In the real world, we don't know what the market really is, and the only thing we have is the belief that the system knows what it is doing. If it doesn't know what even we know, why use such a system? Better to trade with our hands, because we know more than it does and we can handle it better.
wrong message...
 
C-4:

So you deny any at least external "similarity" between the ballooning graph and the price graph?
What does "similarity" have to do with it?