How can I tell the difference between a FOREX chart and a PRNG? - page 8
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
OK, let's say we generate a series with an ACF identical to the real one. What next? Is it possible to earn on the ACF of the real market? I tried it - even without a commission it failed. So the question is - what is the power of this knowledge? We cannot tell the difference between the SB and the market by this indication, but we still cannot make money.
Well, the ACF is not all the characteristics, and who says that we have defined it correctly.
As for earning, that's if the noise level allows it, which is not always the case.
Negra, be jealous. You're the only one who's talking nonsense here. And stop thinking you're gonna get it all on a platter for nothing. If you can't tell the difference, it doesn't mean others have to teach you. Besides, I didn't say I can or can do anything. All I offered was for anyone willing to play the game (which Demi tried to do in the first few pages of this thread). But Demi has consistently shied away from my offer - but that's his right.
To what? Your logic? It's occupied, there's a blonde jealousy there.
PS: What a saucer, theorist. I don't want that.
Negra, be jealous in silence. You're the only one who's talking nonsense around here so far. And stop thinking that you'll be handed everything on a silver platter for nothing. If you can't tell the difference, it doesn't mean others have to teach you. Besides, I didn't say I can or can do anything. All I offered was for anyone willing to play the game (which Demi tried to do in the first few pages of this thread). But Demi has consistently shied away from my offer - but that's his right.
Thanks, but I don't play games - I'm all grown up.
Yeah, well, it's not going to work. The ACF alone is clearly not enough.
A finite amount of data in general is not enough to make such a 100% confident distinction. Especially a small amount.
2 Topekstarter: Your efforts are understandable and well deserved. However: are you sure that you clearly see the statistics that would be sufficient to conditionally identify the two series if they were to conventionally match?
You need to be very rigorous and clear about the task itself here. Only then will your attempts be valid.
I have written before that the criteria for comparing PRNG and real series lie more in the area related to the trading strategies themselves, which directly use these statistics. To put it crudely: for the "Two Machs" strategy there is one set of statistics, and for the "Bollinger + ATR" strategy there are others.
1. There is no way to determine from an arbitrary row that it is exactly SB.
2. There are ways that can classify some rows as non-random (non-SB).
Therefore the problem may be reformulated as follows - find the real ones out of the offered ones. The other ones will be questionable - they cannot be attributed to non-SB, nor can they be attributed to non-SB)))
P.S. But you have to determine in advance what is an SB. Just a lack of memory on the direction? Simply you can generate with volatility memory repetition the real series and then the memory in the will remains, but there is no memory in the direction of the increments. The memory in the ox is not only its intraday cyclic frequency, but also the clustering effect, for example. It shows up on the daily cycles as well.
Avals:
refer to non-random (not SB)
There seems to be obvious confusion among many here about basic concepts like "random series", "random rambling" and "predictability". Just in case, let me remind you:
"Random variable" is absolutely anything whose value cannot be accurately predicted. In other words, even if you are 99.9% sure that tomorrow's potato price will be the same as today's, that value is still random because in principle there is some uncertainty (a probability distribution of possible values).
"Random walk" is a random series (a sequence of random variables) that is formed in a particular way, namely, by summing up successive values of some given random variable (usually having a zero expectation).
By "predictability" (presence of regularities, possibility to earn) we mean the possibility to specify at the predetermined moments (possibly, but not necessarily, even at any moment) a confidence interval, in which the value of the series will fall with an acceptable probability, and (as applied to our cases) such that the expected payoff of the trading system, built on these intervals, is greater than zero almost certainly.
In short, "random" does not mean "unpredictable" and vice versa. Let us use a common terminology.
It seems that there is an obvious confusion of basic concepts such as "random series", "random walk" and "predictability". Just in case, let me remind you:
"Random variable" is absolutely anything whose value cannot be accurately predicted. In other words, even if you are 99.9% sure that tomorrow's potato price will be the same as today's, that value is still random because in principle there is some uncertainty (a probability distribution of possible values).
"Random walk" is a random series (a sequence of random variables) that is formed in a particular way, namely by summing up successive values of some given random variable (usually having a zero expectation).
By "predictability" (existence of patterns, possibility to earn money) we mean the possibility to specify at some predetermined moments (possibly, but not necessarily, even at any moment) a confidence interval, in which the value of series will fall with an acceptable probability, and (as applied to our cases) such that the expected profit of a trading system based on these intervals is greater than zero almost surely.
In short, "random" does not mean "unpredictable" and vice versa. Let us use common terminology.
You insist on talking about this))) http://forum.alpari.ru/showthread.php?t=70386&page=44 post number 439.
Also stubbornly not responding to Demi about the days.
PS: Although dayshare has high-low statistics too. And there are regular bank operations, weekly monthly quarterly.
It's not interesting to talk about the days. Most people trade on smaller timeframes where it's easier to find patterns. Applying statistics to the entire series and reasoning about the similarity of this series to another is silly. In one series there may be patterns that can be used in trading, while the statistics will assert that the autocorrelation function and moments of this series are the same as those of the random walk. So this theorist will prove to himself that one cannot earn on Forex.
It's not interesting to talk about diaries. Most traders trade on smaller timeframes where patterns are easier to find. Applying statistics to the whole series and arguing about the similarity of this series to another is silly. In one series there may be patterns that can be used in trading, while the statistics will say that the autocorrelation function and moments of this series are the same as those of a random walk. So, this theorist will prove to himself that one cannot earn on Forex.
And most of them lose according to statistics. Because they use patterns for trading but not for earning.
Yes, the trouble with these theorists.............................................
But actually the purpose of this thread is different.