From theory to practice - page 475
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
There are even more, incomparably more, Dimitri...
For each order (read - number of summands of exponential numbers) there are also p=from 0...to infinity and q=1-p.
And I say that I am now disintegrated in these flows - and I will integrate in order 60 at p=0.5...
I want to compare my distributions with standard distributions for OPEN and CLOSE for M1.
But in principle, the fact that the market has a memory is enough.
Memory.
Yes, this is the Grail you are looking for. Anyone who knows how to quantify this stuff knows all about the market.
However, you, as far as I know, are using Hearst. Wrong... IMHO.
You should use either ACF or non-gentropy.
I'm working with ACF at the moment - but, no one to even ask for calculation and application details. Prival and Avtomat seem to be the only people here who know a thing or two about it. And they've all left the forum... Only stumps like Unicornis are left... Moron after moron - that's the whole forum... Ugh.
Memory.
Yes, this is the Grail you are looking for. Anyone who knows how to quantify this stuff knows everything about the market.
You have to use either ACF or non-entropy.
Memory is called trend in the right terms. ACF can help extract the cyclical component of a trend, but it is not the whole trend.
The nonentropy and Shannon's theory are out of place here...
Alexander_K2:
Only stumps like Unicornis are left... Moron to moron, moron to moron - that's the whole forum... Ugh.
The more retarded the pattern, the more reliable it is and vice versa. ))
It remains to close the issue of Erlang flows...
So, let's take last week's data on EURUSD from the 60th order Erlang flow (receiving frequency - about once per minute).
We analyze the value of (Ask+Bid)/2.
For the sum of the increments in the sliding daily time window (i.e. for the sliding sample = 1440 values) we have the following histogram for the increments:
Statistics:
We see almost perfect symmetry.
Please post OPEN/CLOSE M1 data for EURUSD for last week in csv format - it is interesting to compare...
It remains to close the issue of Erlang flows...
So, let's take last week's data on EURUSD from the 60th order Erlang flow (receiving frequency - about once per minute).
We analyze the value of (Ask+Bid)/2.
For the sum of the increments in the sliding daily time window (i.e. for the sliding sample = 1440 values) we have the following histogram for the increments:
Statistics:
We see almost perfect symmetry.
Please, send me data of OPEN/CLOSE M1 for EURUSD for the previous week in csv format - it is interesting to compare...
Your sample with no time reference =0.
Your sample with no time reference =0.
If OPEN/CLOSE M1 show the same distribution of increments - I'll give up Erlang flows, so no one else hassles like me :)
If OPEN/CLOSE M1 shows the same distribution of increments - I'll give up Erlang flows so no one else has to suffer like me :)
Try High and Low - these are more adequate BP points and are less random. The more you cling to randomness, the more random the TS will be...
Try High and Low - these are more adequate BP points and are less random. The more you cling to randomness, the more random the TS will be...
OK.
So.
As usual, I checked the data for (Ask+Bid)/2 OPEN M1 EURUSD by myself without any help (I don't expect any - it's like expecting help from children or morons, which are the vast majority here on the forum) according to Dukascopy.
The result:
As we can see - a dramatic difference from Erlang's 60th order flow for the worse.
Both the kurtosis and asymmetry have absolutely the worst result in comparison.
This is crap, not a distribution.
Conclusion: OPEN/CLOSE M1 is impossible to sum up any theoretical basis at all and, consequently, it is impossible to make stable profits.
But Erlang's flows rule.
There you go!