Profit from a random price range - page 4
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
2. Stationarity of the process: probably yes. I don't think the distribution function changes over time either.
1. Too lazy to dig and read now, in view of the last remark:
There is for example a Kolmogorov-Smirnov test, for which, with a random sample, one can test whether the distribution of a random variable is normal or not: https://en.wikipedia.org/wiki/Kolmogorov-Smirnov_test. If that's not enough for you, then please merge all of the above into a description of what you are proposing.
3. yes, even if it's about Box-Muller, there are a lot of different methods. Even here there is a statistical library (from klot, I think), there is a function, the inverse of normal, just for generating a normal value from a uniformly distributed one. In any case, the fundamental transformation law of probabilities is the basis here. That's the law I'm referring to.
As for what I'm missing: I'm not doing it, but only pointing out that this is probably what S.V. wanted to do. He apparently wanted to collect statistics on Returns, and then, based on the empirical distribution of Returns, to transform this data into normally distributed ones, on which, according to his hints andRosh's claims, one can just chop cabbage. In doing so, each dimension of the real Returns will be mutually unambiguously matched by the "normalised" one. On the "normalised" data, trades are opened/closed and transformed into trades on the real data.
1. And you read Peters, there's a lot of interesting stuff in there. I don't need to do the Kolmogorov-Smirnov test to check the normality of Returns, as I know they are not normal, and this is really obvious - for example from the fact that there are heavy tails. Six sigma type events in the real market are quite rare, but still hundreds of thousands of times more frequent than the normal law.
1. you should read Peters, there's a lot of interesting stuff in there.
How about Peters?
Э. Peters "Chaos and order in capital markets"
Э. E. Peters "Fractal Analysis of Financial Markets. Applications of Chaos Theory to Investment and Economics".
2. Stationarity of the process: probably yes. I don't think the distribution function changes over time either.
1. Too lazy to dig and read now, in view of the last remark:
There is for example a Kolmogorov-Smirnov test, for which, with a random sample, one can test whether the distribution of a random variable is normal or not: https://en.wikipedia.org/wiki/Kolmogorov-Smirnov_test. If this is not enough for you, then please merge all of the above into a description of what you are proposing.
3. yes, even if it's about Box-Muller, there are a lot of different methods. Even here there is a statistical library (from klot, I think), there is a function, the inverse of normal, just for generating a normal value from a uniformly distributed one. In any case, the fundamental transformation law of probabilities is the basis here. That's the law I'm referring to.
As for what I'm missing: I'm not doing it, but only pointing out that this is probably what S.V. wanted to do. He apparently wanted to collect statistics on Returns, and then, based on the empirical distribution of Returns, to transform this data into normally distributed ones, on which, according to his hints andRosh's claims, one can just chop cabbage. In doing so, each dimension of the real Returns will be mutually unambiguously matched by the "normalised" one. On the "normalised" data, trades are opened/closed and transformed into trades on the real data.
1. And you read Peters, there's a lot of interesting stuff in there. I don't need to do the Kolmogorov-Smirnov test to check the normality of Returns, as I know they are not normal, and this is really obvious - for example from the fact that there are heavy tails. Six sigma type events are quite rare in the real market, but still hundreds of thousands of times more frequent than the normal law.
3. Do we know that the quantities are evenly distributed? Or in general, what is our distribution function? If so, we have a distribution function that we can transform. Kolmogorov can help here too.
1. Reading the previous description on 1 about stability, it actually duplicates point 2 about stationarity, as far as I understand. About Peters - I'll take and read it, thanks.
About the enterprise itself - let's see what they get up to. If they suddenly disappear here, it's worth taking a closer look.
How about Peters?
1. you should read Peters, there's a lot of interesting stuff in there.
Peters, perhaps?
Э. Peters "Chaos and order in capital markets"
Э. E. Peters "Fractal Analysis of Financial Markets. Applications of Chaos Theory to Investment and Economics".
Thank you, the links work not only in Russia. I am interested in books on money management, can you suggest something? Matemat, the question is also for you :)
Thank you, the links don't only work in Russia. I am interested in books on money management, can you suggest something? Matemat, a question for you too :)
Classics of the genre.
Р. Vince, The Mathematics of Money Management.
For autotrading
Yuri Reshetnikov "MTS and Money Management Methods"
1. Reading the previous description on 1 about stability, it actually duplicates point 2 about stationarity, as far as I understand it.
No, it does not. A stable probability distribution is this (from Shiryaev, vol. 1, p. 232):
Something similar is infinitely divisible distributions.
1. Reading the previous description on 1 about stability, it actually duplicates point 2 about stationarity, as far as I understand it.
No, it does not. A stable probability distribution is this (from Shiryaev, vol. 1, p. 232):
Something similar is the infinitely divisible distributions.
In his time, he was well exposed on some serious forums.
So there is little point in reading his articles, except for the fun of it.
"Dub proved the impossibility of systematic winning on a random series of data"
- This is generally incorrect, unless the random series in question is specified.
For example on such a random series X = a + b*t + e it is very easy to make money (e is a random variable)
There are many other random series on which you can build a system.
The main point is that there are random series with memory and those without memory.
There is a random series with memory; it has a distribution function of increments of a random variable (e) which DOES NOT depend on its previous values.
A random series without memory - its distribution function of increments of a random variable does NOT depend on the previous values of the series.
It is impossible to construct a profitable system on a random series without memory.
As for the normal distribution - the quotes are distributed normally around the moving average, just as S.W. wrote and what lies in the palm of his hand, so we're in the clear here.
1. The kind of distribution function of the differences in price and the mean depends on the variance of that distribution and the value of the mean.
2. The distribution function of this difference is asymmetric, so it cannot be Gaussian.
3. Under certain conditions, the distribution of the difference tends to a Gaussian distribution, but never becomes one.