Why is the normal distribution not normal? - page 5
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
And in other words, if you are strictly picky, the input should not be a number of first differences, but a number of first differences of logarithms (MO=0), or ratios instead of differences (MO=1).
Bo quotient is a ratio C1/C2, i.e. it is multiplicative in origin (C1 * 1/C2).
>> It goes like this:
О! I do have some genius ideas, though. It's more of a parabolu, though. I think I'll write to the nobel committee. Um... Sergei's gonna have to co-write it....
And in other words, if you are strictly picky, the input should not be a number of first differences, but a number of first differences of logarithms (MO=0), or ratios instead of differences (MO=1).
Since quotient is a ratio C1/S2, it is multiplicative by nature (C1 * 1/C2).
This is correct if the price changes by more than 10% on the BP section under study. For exchange rate quotations this is always the case and differences can be used.
Urain, it should be something like this (this is from Peters):
Yen, daily returns over 20 years
Our grandmothers cried forex after all. In 2009, on December 1, the mql forum proved the random wandering of price series. Amen to that...
:) :)
This is correct if the price changes by more than 10% on the BP section under study. For exchange rate quotations this is always the case and differences can be used.
Yes I think I almost agree, however please explain the derivation of the parabola in the way suggested.
??
I've heard many times about thick tails of distribution, but I still do not understand what the point is, I made an indicator which outputs bar size distribution (based on Close[i]-Close[i+1] difference) into separates, can someone explain why distribution of bars is narrower than normal?
The benchmark is a red line yellow histogram.
And the indicator that was used to build it. Original title (Distribution_HGN_&_norm_test)
Why should the measured distribution be close to a normal distribution?
Why do so many people here use a "returns" distribution? It is almost impossible to use it afterwards. What good is it if it resembles normal and stationary.
Why isn't the price distribution used? After all, that's the main thing that's interesting.
Why does the measured distribution have to be close to a normal distribution?
Why do so many people here use a "returns" distribution? It is almost impossible to use it afterwards. What good is it if it resembles the normal and stationary distribution?
Why isn't the price distribution used? After all, this is the main thing that is interesting.
The price series is not stationary.
That is, its expectation is known only in Sochi. That's where they use price distribution. Only they don't write about the results here, the bastards.
// Make the travel arrangements. Tell us about it later.
In other cities, as well as in our rural areas, they are content with what it costs - the first differences.
Why does the measured distribution have to be close to a normal distribution?
Why do so many people here use a "returns" distribution? It's almost impossible to use it afterwards. What good is it if it resembles normal and stationary.
Why isn't the price distribution used? After all, this is the main thing that is interesting.
In my opinion the price is identical to its first difference and fully recoverable from it, so whatever is convenient for the study is used,
Imho there is no useful information in the cumulative sum (price) distribution at all.