On the unequal probability of a price move up or down - page 114
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
Mikhael1983:
"... Fools can go on believing in something like...":
It's time to set the impetus for the weekend's discussions...
Although I have tried to say little, I have given the minimum (the very minimum) of what is necessary for those with brains to start THINKING.
However, perhaps a slightly larger stratum than the smartest have not been able to make ANY independent moves, so I will show that even seemingly BANAL things can turn out very unexpectedly.
Let me remind you:
Now for the best part: the volatility and shape of the EPvirt virtual cross differs from the volatility and shape of the real EP cross.Or, even more strongly: the volatility and shape of one EPvirt1 virtual cross (set by some alpha1 value) is different from the volatility and shape of another EPvirt2 virtual cross (set by some other alpha2 value).
Think )
Why should we think, you've made a "throwing stuff on a fan" and want to engage our bright, uncluttered minds in solving your genius scheme? )))
You'd better think about it yourself and tell us about it. That would be great.
And why should we think, you've made a "blast on the fan" and want to engage our bright, uncluttered minds in solving your brilliant scheme? )))
You'd better think about it yourself and tell us about it. That would be great.
Maha, try to get away from the formula: "Me and everyone else" and life will get better.
Maha, try to get away from the formula: "Me and everyone else" and life will get better.
I seem to have everything in order )). The world certainly doesn't revolve around me, if that's what you mean.
But what do you say if I introduce new "virtual currency pairs" as follows: EDnew = ED + alpha, PDnew = PD + alpha, where alpha is some constant.
They may be traded directly. Because their increments are identical to the ED and PD increments. By definition. For the derivative of a constant is zero.
But this directly means that one can trade them "crosswise":
Once again slowly: this "virtual cross" CANNOT be traded UNLESS (it's not in the terminal), but CAN be traded by decomposing into ED and PD trades.
And now for the best part: the volatility and shape of the virtual cross EPvirt differ from the volatility and shape of the real cross EP.
Or, even more strongly: volatility and form of one EPvirt1 virtual cross (given by some alpha1 value) differs from volatility and form of another EPvirt2 virtual cross (given by some alpha2 value).
Think )
Yep, and then you yourself cite charts that don't differ in either shape or volatility on a scale. Nothing can change from the addition of a meaningless constant that should be reduced for lack of use. What happens is that the changing values of the ratio shift to the right of the comma, that's all. Well, that's a good start, why make it so fat? You should be more creative.
Yep, and then you yourself provide graphs that are neither shaped nor volatile in scale. Nothing can change by adding a meaningless constant, which should be reduced for lack of use. What happens is that the changing values of the ratio shift to the right of the comma, that's all. Well, that's a good start, why make it so fat? You could have been more creative.
It looks like he copied it from someone else and didn't get it himself.
;)
Yep, and then you yourself provide graphs that are neither shaped nor volatile in scale. Nothing can change by adding a meaningless constant, which should be reduced for lack of use. What happens is that the changing values of the ratio shift to the right of the comma, that's all. Well, that's a good start, why make it so fat? You should be more creative.
It's a priori obvious for sane (in contrast to fools) people that ED/PD and (ED+alpha)/(PD+alpha) have different volatilities and shapes.
I will draw them together, for illustration (for what I will move the virtual cross downwards and add to it the value of beta = -0.068)
Also it is shown, that Pearson's coefficient of linear correlation of these two plots on this interval of time samples is less than unity (it cannot be otherwise).
Looks like he cheated off someone and didn't get it himself.
;)
It is a priori obvious for reasonable thinkers (unlike fools) that ED/PD and (ED+alpha)/(PD+alpha) have different volatilities and shapes.
I will draw them together, for illustration (for which I will move the virtual cross downwards and add to it the value beta = -0.0685).
Also it is shown, that the linear correlation coefficient of K. Pearson of these two charts on this interval of time samples is less than unity (it can't be otherwise).
fools and fools think alike (popular proverb - I emphasize that I do not mean to offend anyone personally, and I do not want to offend anyone)remind me, what was your last combination that went down - pound to sell, eva to buy?
I'm just checking, it seems to work the same way for me
I don't play with lots like that.
There's a mistake there.
You do the math, what would happen if the lots were the same
It is also shown that the K. Pearson linear correlation coefficient of the two graphs is less than unity at this time interval (it cannot be otherwise).
If you want to invent statistical arbitrage, correlation has nothing to do with it. Two strongly correlated instruments can diverge for as long as you like. What matters here is cointegration, not correlation. Read for example here. And compare returns at high correlation and cointegration. Correlation and cointegration are not related and independent concepts.
What the author is right about is that maths is probably hard to push on a forum. It's understandable, since not everyone has a degree in maths, physics, MIT, etc:)
Basically:
p.1 - everything in the last two posts from the author is correct (you can. who wants, yourself in excel, take your word for it or ask someone with appropriate education);
a*EURUSD-b*GBPUSD synthetic in different combinations of a and b allows to trade pairs separately (a>>b or a<<b) and their cross and their difference and an infinite number of cross modifications.
p.2 - so far (at least for me) it is not clear how it helps to obtain a practically guaranteed profit in a triangle.
in particular, with lots a>b almost three times the eu will make a downward walk to close in minus what is called "by definition":)
apparently the estimated lots change during the day slowly, and say over the week - already substantially;
by the way, the correlation of pairs was awful at the time of entry on M5 576 - minus 0,24, maybe it affects the result.
P.S. The triangle representation through ND and EN (new basis) is also interesting, by the way, but for now p.2.:)
P.P.S. Star arbitration even with 7 pairs and moreover with two does not work - this information is somewhere 5 years, or even 10 years old (when hrenfx wrote his toys and everybody played with them + wrote his own).