Not the Grail, just a regular one - Bablokos!!! - page 28
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
Alexander, are these your words?
" Now I hope that the number of prosperous over a long period of time will increase many times over. That's not insignificant either! :) And this % will grow many times over, if traders who read this thread, will not listen to anyone, and take for themselves to check the methodology (based on Hedging), which I offered above in this thread. People! Take an example from the funds, as they trade, and you will be happy. :) One of the techniques, based on the principle used by the funds and banks, I described above. You can speculate and trade at the same level as the chubby ones.
nope - not mine :-) you what... I'm a miser... and selfish... it gives me the creeps to think that other traders can make a steady income...
that's my loss :-) because my profit is the loss of other traders... at their expense the office pays me my winnings regularly... and to breed competitors... not for me :-)
so... a little half a hint - for those who have a working pumpkin I can still do something ... but otherwise ... grail and to the Masses - FUCK YOU... suck it :-)
homo homini lupus est
I was thinking about how to compare, say, 2 pairs with each other or a simple hedge triangle.
how to do it, because correlation does not help us here, because points of similarity are not placed synchronously and linearly, it should be correlation with a shift in time, and also with a change of different depths of the sliding window....
HOW else to compare 2 series. if similarity points are scattered non-linearly.
Maybe we need something comparative, not based on correlation of series, but on comparison of distributions of series, degree of similarity of distributions, or "correlation of distributions " with a sliding window and with a sliding growing window.
this may be a useful (but probably not) idea of Zarembka's method
http://www.aup.ru/books/m157/3_4_1.htm
Well, I haven't given up the way of earning money on SB, but when calculating indexes the degree of influence of pairs is different https://www.mql5.com/ru/forum/117367/page6 the bottom post with a picture. that's what I wanted to add to it.
The question is how to compare the similarity of the distributions, probably will determine a certain mass of instruments in the overall pile.
The question is HOW TO COMPARISE 2 DISTRIBUTIONS??? by which criteria, similarity, excesses or some other, you should think it over.
By the way, it is probably better to draw distributions of equal-volume bars than to draw distributions of simple bars.
The form of the distribution should be compared, not the absolute values, because different pairs have different volatility and different trade volumes.
This is why the distributions (sizes of distributions) will be very different, but what is needed is not this difference, but something comparable by degree of similarity.
Nonparametric statistics and distribution fitting
I'll continue with the text... here comes the month of trading ending (not from the beginning, from the middle) 15 days of trading have passed....
xxxxx4441 d15 t15 buy 0.20 p1 1.3338 0.0000 0.0000 d15 t15.1 1.3325 0.00 0.00 0.00 -26.00
xxxxxx4451 d15 t15 sell 0.59 p2 1.5770 0.0000 0.0000 d15 t15.1 1.5756 0.00 0.00 0.00 -82.60
xxxxx4454 d15 t15 sell 0.42 p3 0.9815 0.0000 0.0000 d15 t15.1 0.9783 0.00 0.00 0.00 134.40
xxxxx4455 d15 t15 sell 0.51 p4 0.9958 0.0000 0.0000 d15 t15.1 0.9967 0.00 0.00 0.00 -46.05
trades closed the same day, 8 hours after opening :-)
It's better to search for pairs by indices, and lots are a thing of the past.
7-8 rows is still better than 32-64 of their "/" ratios.
Thanks for the puzzle).
Since the month in the State is over, we can sum up a little...
From an initial depo of 1500 cu, the depo has grown to 3660.73 cu based on transactions... almost 100% in a month of trading...
Tomorrow we will go to the next month, but will not increase the depo load :-) Let's see what and how it changed during the last month ... what pairs to trade with :)
price is divided by time counts, or by movements in pips. The distribution is naturally not normal. But if we connect all consecutive points of the series and then erase all of them, we will obtain a curve on which we can again draw (adjust) readings so that the distribution becomes normal; then perhaps analysis of this very series will be more informative, comparing such transformations with other pairs.
There are also 2 - 3 .... There are also 2/395 measure normal distributions.