Thoughts on some of the absurdity of multi-currency analysis. - page 18

New comment
 
It's complicated, it can't be explained on the fingers. And there isn't much useful information. I plan to re-calculate the dimensions in some time, and I will present the results then. I can tell those who are keen on making money that there is no direct link to "how will this help us raise money?
 
The dimensionality in this case is something like correlation, only if correlation always shows a linear relationship, then the Fr. dimensionality will show any relationship, including non-linear.
 
has the topic descended into abstract research?
I believe that "multi-currency analysis" itself for the purpose of predicting currency pair movements is simply not possible, because the movement of all currency pairs begins after at least one pair changes.
But analysis of the strength (amplitude) of currency pair changes or the duration of currency pair fluctuations, i.e. time frame analysis makes sense. If you can predict at least the duration of a short-term trend, you can have a good chance of making a successful transaction even without knowing in which direction the currency pair will move.
 
you have an open long position in NZDUSD
you learn from CNN about Greece's budget problems
you understand that Greece's budget problems will negatively affect EURUSD and you know the positive correlation between EURUSD and NZDUSD = you are closing your long position
is this a multi-currency analysis?
 
Demi >>:
у вас открыта длинная позиция по NZDUSD
вы узначете из выпуска CNN про бюджетные проблемы Греции
вам понятно, что бюджетные проблемы Греции негативно скажутся на курсе EURUSD и вы знаете про положительную корреляция между EURUSD и NZDUSD = вы закрываете свою длинную позицию
это мультивалютный анализ?


have you ever thought about the "economic" influence of Greece? how a country like Greece can have an impact on the euro??? has anyone heard about minerals in Greece, about goods produced in Greece??? 90% of forex news are just news that cannot change the course of currency pairs, to excite for a couple of hours - yes, I agree, but countries like Greece cannot radically change the trend. As the saying goes: you are "looking for black cats missing in a dark room". Well, the correlation of currency pairs is not visible on all timeframes
 
Demi >>:
у вас открыта длинная позиция по NZDUSD
вы узначете из выпуска CNN про бюджетные проблемы Греции
вам понятно, что бюджетные проблемы Греции негативно скажутся на курсе EURUSD и вы знаете про положительную корреляция между EURUSD и NZDUSD = вы закрываете свою длинную позицию
это мультивалютный анализ?

The positive correlation here says nothing.

The same currency, the quid, is part of the pair. That is, the currencies themselves (Euro and NZ) may not depend on each other in any way. Correlations must be used properly. The fall of the Eurobucks does not have to affect the NZDbucks. In your case you are only betting that the quid will rise relative to 2 currencies at once.

 
1. I don't understand what "Greece's economic clout" is. I understand that the market perceives Greece as an indicator of the general state of the eurozone, that the market perceives Greece as the first swallow of troubled countries like Spain, Portugal, Italy and the UK. That is my understanding. I also understand that the eurozone is a complex system and the performance of any system is determined by the performance of its slowest component, which at the moment is Greece.
2. Tell me about the minerals of Japan and the goods produced by Switzerland.
3. The forex market mostly operates with emotional news rather than economic ones. that's why the reaction of price movements is often completely inconsistent with the economic component.
4. Please explain to me the price movement from 04.12.2010 to today from 1.5141 to 1.3282 (most market participants explain this by problems in Greece etc.)

5. The correlation of currency pairs is visible at all intervals and at all times. It changes over time, can change its sign to the opposite, but never equals zero.

???

 
a strong euro movement is necessarily reflected in the NZD, there is not just a correlation, there is a strong correlation
the other thing is that the fall is not proportional
that was an example
replace the NZD with the Fund
the correlation is even stronger there
 
You see, Demi, are you trying to prove that the news about Greece (and other small countries) has set the Eurobucks downward trend, or maybe the dollar is just strengthening?
Well, if the Forex market operates only on emotions, then it is completely unrelated to the economic model :)

Well, about the correlation - the correlation is always calculated based on already existing data and conclude how similar the behaviour of random variables was at a particular time interval. I don't know if the correlation is calculated for future movements of currency pairs :) .Correlation is a fitting of existing values to the time intervals. I, by the way, wrote four posts above - I believe that the movement of any currency pair causes the movement of other currency pairs - that's correlation for you.
 
the downward trend in the Eurobucks may indeed have caused the head shaman of the island of Papua New Guinea to have a runny nose
and there may well be 854 other reasons which we do not know about, which "in fact" caused the downward trend.
But using Occam's razor, it's still more likely that it was Greece that crapped.
As for your point about the movement, I agree 100%.
but there is a VERY LARGE probability that if during the n-last period the correlation between two currencies was e.g. +0.82, tomorrow it will not be -0.25, and if it changes, it will be within a very narrow band
Of course, we do not take into account such extreme cases as:
1. market noise, e.g. correlation on 1-5 min timeframes
2. Release moments of super important macro news for the currencies in question.

ok?
1...111213141516171819202122232425...40
New comment