Absolute courses - page 30

 
Dr.F.:


P.P.S2. And yes, the question of correlation and the question of orthogonality, dependence (functional, stochastic, or otherwise) are in no way related. Typical example: sine and cosine. Correlation is zero. Dependence is strictly functional. Learn the math.

Oh, really? Are you going to poke me with a freshman classic?

I never once used the word dependence, I only said that the above condition on correlation is incorrect and should be replaced by the opposite.

 
You have used the word orthogonality. Orthogonality means independence. Otherwise it would not be a set of basis (functions, vectors, not the point). My condition is correct and reflects the desired physics, which I have already explained. So there is both reasonable hypothesis and mathematics. Next, trade. But first a couple more days of refining the mathematics, bringing it to the absolute.
 
Dr.F.:

Fractions of a percent? Per year? Right, right. Dollar inflation is 2% a year, the ruble 8-10% a year And so for 10+ years. One value becomes cheaper at a rate of (allegedly) 2% per year, the other at a rate of (allegedly) 8-10% per year. The ratio between the two over a 10+ year interval is almost constant. STRANGE, ISN'T IT? What if I told you that real inflation is 15-20-30% per year in both?


Have you ever heard of deflation in the euro zone, in Japan? It's when inflation is <0, an extremely negative phenomenon, however, it has been fought there.
 
Dr.F.:

You have used the word orthogonality. Orthogonality means independence.

If anything, orthogonality in mathematics is a general concept which means that some generalised quantity, which can be called a scalar product, is equal to zero. In this case, the correlation coefficient is zero.
 
alsu:

Have you ever heard of deflation in the euro zone, in Japan? It's when inflation is <0, an extremely negative phenomenon, but it's been fought there.

Save the fairy tales for the idiots. Build the yen's exchange rate against gold, not against the dollar. And also the S&P 500 index against gold. You'll have a partial epiphany. The reality is even harder, because not only is gold not constant, but it has been depreciating all this time compared to the real wealth of the real physical economy. Look at the price of rice on the world market in Yen or rather the price of Yen in grams of rice. And show me deflation. Deflation means that EVERYBODY around here has earned NOTHING. Ha. Any idiots out there who believe that? That's why governments print their "legitimate money", so that only it can make money, by legally taking money from the users of their "legitimate money". The fact that this "tax" was less in the yen zone at some time than in the dollar zone does not change its sign. It cannot, by definition, be <0.
 
Dr.F.:

Once you have grasped the NATURE of changes in E/D, E/Y, D/Y RELATIONSHIPS, that is, if you separate the GENERAL (approximately the average of forms E, D, Y) and the INDIVIDUAL (forming ED, EY, DY) evolution of forms E, D, Y, you will naturally move to a stable trading system based on the preponderance of probability: You will open trades with equal TPs and SLs in the direction (e.g. buying EURUSD) in which the Eurodollar is moved by BOTH processes (both the evolution of the Euro and the evolution of the dollar). For the probability of them both changing sign or ending at once is less than 50%.


I.e. both E up and D down (then E/D up)? By your construction they are correlated with values close to 1. How will they go in different directions?

P.S. alsu has already asked))

 
alsu:
For that matter, orthogonality in mathematics is a general concept which means that some generalised quantity, which can be called a scalar product, is zero. In this case, the correlation coefficient is zero.



This "generalised quantity" (scalar product) cannot be the correlation coefficient. It has no right. Learn the math.
 
Avals:


i.e. both E up and D down (then E/D up)? You have them correlated by construction with values close to 1. How will they go in different directions?


How about a closer look? Linear regression to the rescue. No one is saying that this effect is great. It is that this divergence is a small correction to the co-movement of the same. But it is what determines EVERYTHING. And it is the fact that the correlations in pairs are VERY close to 1, but not 1 that gives the right to this divergence. There is a limit to getting even closer to 1, an attempt to go even higher would destroy the relationships with the known (realised in kind on metatrader screens) relationships.
 
Dr.F.:

Deflation means that everyone around us has made money by doing NOTHING.
Deflation means that the domestic producers of most goods cannot sell them due to lower demand and are forced to reduce prices, forcing the central bank to reduce the money supply (this can be done in various ways - by selling securities, increasing the amount of required reserves, etc.).
 
Dr.F.:
Fractions of a percent? Per year? Right, right. Dollar inflation is 2% a year, the ruble 8-10% a year And so for 10+ years. One value becomes cheaper at a rate of (allegedly) 2% per year, the other at a rate of (allegedly) 8-10% per year. The ratio between the two over a 10+ year interval is almost constant. STRANGE, ISN'T IT?


it only means that the rouble is strengthening against the quid.