The most banal trading strategy - page 38

 
Yuriy Asaulenko:

By the way, if you add absolutely predictable,Maxim Kuznetsov is right.

Wrong, even in this case. I, of course, understand what you are saying. What you are saying is that if after some point in time the readings of the price are a function of the readings that preceded that point, it would seem that there is no new information. But in reality there isn't, because the price didn't have to follow the prediction. In this case, the fact that the real price evolution coincided exactly with the forecast is just a coincidence, in reality these counts are still new and independent of the previous ones.

Let's take a look at my forecast for the Eurodollar and see how it plays out. And here's how:

 
mikhael1983isakov:
Wrong, even in this case. I certainly understand what you are trying to say. What you are saying is that if after some point in time the price readings are a function of the readings before that point, it would seem that there is no new information. But in reality there isn't, because the price didn't have to follow the prediction. The fact that the real price evolution coincided exactly with the forecast is just a coincidence, in reality these counts are still new, independent of the previous ones.

In statistical terminology, the point remains the same. A more or less reliable forecast is possible in those BP areas where the amount of new information in the forecast interval is minimal. If possible.)

 
Yuriy Asaulenko:

In statistical terminology, the point remains the same. A more or less reliable forecast is possible in those BP areas where the amount of new information in the forecast interval is minimal. If possible.)

The amount of new information in the forecast interval is always the same (if the forecast interval length is fixed). In short, it is proportional to the number of new readings to be made in this interval (we mean equidistant time grid).
 
mikhael1983isakov:
The amount of new information on a forecast interval is always the same (given a fixed length of forecast interval). In short, it is proportional to the number of new counts to be made in that interval (we mean an equidistant time grid).

I'm not going to argue, otherwise we'll get bogged down).

 
Yuriy Asaulenko:

Actually, if you add absolutely predictable

and there's a long mockery of schoolchildren on the subject of "what is information" :-)

"absolutely accurately" predictable is not necessary - because accurate indicators are not realized in system states.

Our young friend confirms it too, his SMA forecasts aren't exact, but "roughly as it will be". That is, in fact he predicts not the rate itself, but the state of his own machine.

---

By the way, it is from the category of paradoxes of probability and information theories = an automaton with a finite memory size cannot trade profitably on the exchange. To trade profitably its informational capacity must be no less than all others combined.

 
Maxim Kuznetsov:

This is a paradox of the probability and information theories = an automaton with a finite memory capacity cannot trade profitably in the stock market. In order to trade profitably its information capacity must be no less than that of all others combined.

My God, how many empty meaningless terms people invented for themselves. "Information capacity of an automaton"... brrr... and behind it all, nothing, a ringing emptiness.
 
Maxim Kuznetsov:

This is a paradox of the probability and information theories = an automaton with a finite memory capacity cannot trade profitably in the stock market. In order to trade profitably its information capacity must be no less than all others combined.

This is an open question. I, on the other hand, believe it can. Let's say for a while. But, I suppose no one needs an eternal needle for a primus either.

 
Yuriy Asaulenko:

That's an open question. I, on the other hand, believe it can. Let's say for a while. But I suppose nobody wants an eternal primus needle.

Really?! Every third topic here is about the eternal primus...)

even this one...

without understanding the impossibility of solving the issue in a general way, people don't try to look for private/edge, near-crisis solutions, work in the market.

And keep looking for the coveted grail, the general formula, the universal theory, the ultimate extrapolator, the non-collapsing robot and so on.

 

Maxim Kuznetsov:

And they keep looking for the coveted grail, the general formula, the universal theory, the ultimate extrapolator, the non-clearing robot and so on.

Let's see how the EURUSD price evolves and how well it follows the forecast:

 
mikhael1983isakov:

Let's see how the EURUSD price evolves and whether it follows the forecast:

A single forecast, either coinciding or not coinciding with the real one, is nothing at all. If we are speaking about the TF 5m, then in order to confirm it we need some statistics about forecasts for a year and a half. At least on history.

No, I'm not asking you to confirm anything here.