From theory to practice - page 554
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
I wrote this to them about 200 pages ago) well, it will come in 5 years)
it will only work in the stock market - a return to the mean.
they are peeing forex, there is no clearing.
what you described, right:
(0,99067+0,99263) * 50000 = 99165
And how did you calculate the increments?
in the table bid and ask next to each other, enter in the same file the formula for the difference between them and in half, the picture will become clearYes I counted everything there correctly, everything is now as clear as day and night.
I have calculated everything correctly, everything is as clear as day and night now.
bid price increment/movement equal to spread - maximum amount
why are you surprised - i do not understand it yet
so in one of the first posts preceding this branch, I wrote - the probability is equal to 0.5 on + / - 1 point of 4-digits of the current
K2 claimed - 0.05
i could have stopped there, but no ...
this is a hot topic
;)
No, that's right - the sharp peak at zero is pseudo-quotes. The task was just to move to such an Erlang flow so that these pseudo-quotes would not be there and that is where the purest Laplace "sits". But, only - what to do with it? The question is open.
Alexander, as they say, until you do it yourself you won't understand. The doc was right, you give out good pies in your shop, they are straight and beautiful and it's a great pleasure to eat them, while in other places they are crooked and tasteless, just a waste of money. The logarithmic distribution was a pain in the ass, but now it's the exception, not the rule. I will of course look through your files tomorrow, but everything is clear now, exponential readout and alignment by Erlang flows is way to nowhere, do you really believe, that process you want to break is really so deterministic? I'm just once more convinced of that, in principle I've already said it before, exponent on exponent won't do any good. There's another way to approach it.
This will only work on the stock exchange - a return to the average.
they are peeling the forex, there is no clearing.
it's a hot topic.
:) Why shouldn't it live on?
Channel trading is an intuitive thing, and distributions are mysterious. Memory, tails...
How do the classics say it? In memory: "Whoever knows this mystery will never go into a factory again".
Amen.
No, that's right - the sharp peak at zero is pseudo-quotes. The task was just to move to such an Erlang flow so that these pseudo-quotes would not be there and that is where the purest Laplace "sits". But, only - what to do with it? The question is open.
Here is what we have got, when the peak is removed, it all does not look very good. It's all inside the window, because we make the decision inside the sliding window. I am interested to look at the last variants, where you got closer to the binomial, will you let me calculate?
This is what happens when the peak is removed, it doesn't look good. All this is inside the window, because we are making a decision inside the sliding window. I am interested to see the last variants, where you got closer to the binomial, will you let me calculate?
No, that's not it.
I have to put "1" in the outermost column after a certain number of rows in Excel and do sorting. It's like emulating Erlang flows. Have you done that?
It should work nicely.
My order of 300 was strictly kurtosis=3, asymmetry=0.
No - that's not it.
You have to put "1" in the outermost column after a certain number of rows in Excel and do sorting. Kind of emulate Erlang flows. Have you done that?
It should work nicely.
My 300th order was strictly kurtosis=3, asymmetry=0.
Here I didn't take Erlang at all, it's a clean version, right after exponential readout. In this case the window of 10000 for the 300th order is not enough.
I didn't take Erlang here at all, it's a clean option, right after the exponential readout. In this case a window of 10000 for 300 order is not enough.
Nah, the clean option isn't it.
Ahem... I repeat my question - so you found pure Laplace distribution and what? Look up Laplace motion on the internet - same jumps, same junk.
A Bas, who obviously graduated from a rural school without C, assures that going beyond any sigma does not guarantee a return to the mean.
Only exponentially decreasing ACF provides this guarantee.
I.e. among other things it is necessary to transform data so that this condition is always satisfied.
Sorcerer is right - the most important thing in Forex is preprocessing of input data. That's all.