From theory to practice - page 131
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
Back to the previously discussed topic of how to read tick data.
I reread Feynman's works again, read a bit of the works of a certain William Gunn and some interesting threads here on the forum (all of them dated around 2009-2011. - Then it's like a dip in research).
All the same, the current (measured) price value should be read at a certain interval and nothing else - no reading of every tick ! It is from this concept of price observation that the Fokker-Planck equation, and the proportional dependence of price deviation on time via the square root, which Gunn liked to twist into his mystical works (without guessing about Einstein's formula for Brownian motion) and the actual rate of increments, which Feynman operated on, are both derived.
That is all, I do not come back to this question.
Regards,
Alexander_K
PS Those traders in whom TS is adjusted to work with every tick, instantly run into the Heisenberg uncertainty principle and sooner or later will remain with an empty purse and glass eyes from perplexity and horror before an impending poverty.Reread Feynman's work again, read a little of the work of one William Gunn and some interestingthreads here on the forum...
I encourage everyone to take a good look athttps://www.mql5.com/ru/forum/134424
There a man called Farnsworth went even further in his research than I did. No one listened to him and they should have. The guy did a little publicity for a while and left the forum. What a pity!
So, I address this Farnsworth through the years and distances - No, uncle, not the tails of the distribution of increments are the "memory" of the process, but the second of the multipliers of the product of the non-standardized Student's t2-distribution and some exponent forming this very distribution of increments. This exponent that fits the tails of the distribution is the "memory" of the non-Markovian process. So! But I think you've got there yourself, otherwise you'd still be on this forum reading the opuses of ignoramuses. Sincerely, Alexander_K
I encourage everyone to take a good look athttps://www.mql5.com/ru/forum/134424
There a man called Farnsworth went even further in his research than I did. No one listened to him and they should have. The guy did a little publicity for a while and left the forum. What a pity!
So, I address this Farnsworth through the years and distances - No, uncle, not the tails of the distribution of increments are the "memory" of the process, but the second of the multipliers of the product of the non-standardized Student's t2-distribution and some exponent forming this very distribution of increments. This exponent that fits the tails of the distribution is the "memory" of the non-Markovian process. So! But I think you've got there yourself, otherwise you'd still be on this forum reading the opuses of ignoramuses. Sincerely, Alexander_K.
all the "old-timers" checked it out... but it seems that reading it to the end is pointless, no practical conclusions and implementations
p.s. yes, that's how it turned out )
So all the "old-timers" have checked in... but it seems that reading it to the end is pointless, no practical conclusions and implementations
p.s. yeah, that's what happened )
One of the coolest threads ever! With pleasure I read it until the suspiciously familiar faces appeared on it, after which the thread stalled, and the topicstarter generally fled the forum without looking back :)))
Output the densities from the different areas in the column. The samples are the same.
I wonder if the trend with exp. time is strongly maintained or not.
Output the densities from the different plots in the bar. The samples are the same.
I wonder if the trend with exp. time is strongly maintained or not.
Hm... Fascinating assignments coming from you... With exponential time? With exponential time intervals between quotes, maybe? Be a bit more specific, please. Your charts are nice - this is the movement of the wave pack ( probability density function) of the price. Where did you get them from? However, I guess that you have already solved this problem, and now look how I do it. Look - I do not feel sorry. I work for everybody. I am bored working for myself.
By the way, I have 5 trades so far (+3/-2). At the moment profit for 2 days =+269 pips
Prepare your pockets, my friends. Clean them from dust. I'll be sure to post the solution algorithm on the forum in the form of a model.Output the densities from the different areas in the column. The samples are the same.
I wonder if the trend with exp. time is strongly maintained or not.
Ha!
I reread the task again - no,Vizard_, you haven't solved this problem yet, since you ask such questions. Farnsworth gaveyou the answer in Russian in the form of histograms. When did he see it? That's right - around 500 bars in the sample. 500 bars (480 to be exact) - here is the necessary sample to see the whole trend (at M1 and M15, etc. because of the market fractality), and then the memory dims it and collects all the stuff in one pack. That's why it stops seeing the splitting of the packet into two derivatives as the sample increases. Only Farnsworth's hands trembled from happiness and complete misunderstanding of what is going on (simply put - complete stupidity :))) and he could not finish his work :)))
Whether the time is exponential or uniform is not the point. I'm making it a little easier for myself with the exponential scale. Now I don't have to count the averaged archival historical variance and other NE moments for a non-Markovian process. Now it's as if I have a Markovian process with pseudo-states and just use the free run length as the variance. This is correct, but not the whole truth, naturally. It's a bit more complicated there - but it's not for me to teach you, so I'll shut up.
At 6:05 Moscow time my TS had 6 trades in total (+4/-2). Profit = +415 pips
I can't get enough of these 2 negative trades... Shame - what can I say... I even know the reason - I`m using SMA now, and it`s not right. I must switch to WMA and use just time as scales. I'm off to Hilbert space now, but will definitely be back. See you soon!Hi!
What are you guys talking about?
I don't understand anything.