Pair trading and multicurrency arbitrage. The showdown. - page 108
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
100% correlation does not allow you to make money, and 80% also, because it means that pairs diverge every period of time by 20%, i.e. unlimitedly away from each other by 20%, more and more, and never come back together
Pairs can diverge at 99% correlation. And they may not diverge at a correlation of less than 10%.
Co-integration is important for them not to diverge.
Pairs can diverge even at a correlation of 99%. And they may not diverge at a correlation of less than 10%.
Co-integration is important for them not to diverge.
Somebody here said that the euro "is about to break upwards", and I don't mind it in principle....
but restoring old tools with new colours I couldn't help but notice: on EURUSD there is a classic pattern "Three Indians", beautiful, beautiful, just like in TA manuals and illustrations.
and the pattern says that the upward move has failed and the downward move will take place.
We've got a whole week ahead of us, so we'll see.
someone was saying that the euro was "about to go up",
My instruments show the overheating of the euro (not a currency pair, but a currency).
It is very likely that within the next week or two we will see a quick and
relatively deep collapse of the euro.
The situation with the yen is the opposite.
We should get ready to sell EURJPY, but there is no signal yet.
The terms "overbought/oversold" do not quite correctly reflect
essence, but somewhere close.
Pairs can diverge even at a correlation of 99%. And they may not diverge with correlation less than 10%.
Co-integration is important for them not to diverge.
And there is no cointegration in Forex
It does not follow at all that cointegration can be replaced by correlation.
These are completely different metrics, although the names are somewhat similar. )))
It does not necessarily follow that cointegration can be replaced by correlation
For example, Clive Granger and Michio Hatanaka participated in a project on the use of harmonic analysis in economic data as associates of John Tukey (author of the concepts of software and bit). In 1964 Granger and Hatanaka published the results of their research in the best-selling book Spectral Analysis of Economic Time Series. Granger also wrote an article on the same results, "The Typical Spectral Form of an Economic Variable", which appeared a little later in the prestigious journal Econometrica. Both publications provided significant support for the advancement of new techniques in practice.
In 1969, Granger, in the same journal Econometrica, put forward a concept that later became known as Granger causality. The idea of the proposed approach had already been expressed earlier, in 1956 by Norbert Wiener, and was that if taking into account data about one signal helps to predict the behaviour of another signal, it may mean that the process generating the first signal influences the process generating the second signal.
https://www.mathnet.ru/links/8ebd6927d22de40ef05e0d5087b75137/ivp345.pdf
and now DSPs are running around the world with this co-integration like a piece of paper....For example, Clive Granger and Michio Hatanaka participated in a project on the use of harmonic analysis in economic data as associates of John Tukey (author of the concepts of software and bit). In 1964 Granger and Hatanaka published the results of their research in the best-selling book Spectral Analysis of Economic Time Series. Granger also wrote an article on the same results, "The Typical Spectral Form of an Economic Variable", which appeared a little later in the prestigious journal Econometrica. Both publications provided significant support for the advancement of new techniques in practice.
In 1969, Granger, in the same journal Econometrica, put forward a concept that later became known as Granger causality. The idea of the proposed approach had already been expressed earlier, in 1956 by Norbert Wiener, and was that if taking into account data about one signal helps to predict the behaviour of another signal, it may mean that the process generating the first signal influences the process generating the second signal.
https://www.mathnet.ru/links/8ebd6927d22de40ef05e0d5087b75137/ivp345.pdf
and now DSPs are running around the world with this cointegration.