Alternative and common approaches in the construction of TC - page 9
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
FGBMH1 - against (FGBLH1+GFBSH1)
I would only assume such a thing after the test has been run at various values of the parameters mentioned. It's entirely possible that other laws might work for compliance (if they are among those tested, of course, but what if they aren't?). That's the problem.
It's not a problem. The point is simply that the initial instruments are tested exactly for the exponential distribution at the 0.00001 significance level, i.e. one thousandth of one percent. When we form a synthetic distribution, this distribution will be blurred only at the top because small deviations correspond more to market noise and it is here where the instruments have the maximum degree of independence between each other which allows us to apply the DPT and assert that in a limited area near zero the synthetic will be distributed near to the normal law and the greater the number of initial FI, the closer. But on the tails the situation is opposite - broad market movements are very strongly correlated between different FI, as they are caused, as a rule, by fundamental reasons. As a consequence, in the area of large deviations the independence condition is not fulfilled and the TPT does not work: the exponential tails remain as they were.
So if you try to identify the synthetic distribution as any of the known ones, it probably won't work, as it is rather some mix of two different laws.
Nevertheless, my request to hrenfx stands - to give an example of a synthetic instrument that is supposed to have significantly different characteristics from individual FIs. Purely for research purposes, just to see if the performance improvement actually exists or is a sham.
Nevertheless, my request to hrenfx still stands - to give an example of a synthetic instrument which is supposed to have significantly different performance from individual FIs. Purely for research purposes, just to see if the performance improvement actually exists or is a sham.
He gave an example. True, on a short stretch of history, yes. =)
It's not a problem. The point is simply that the initial instruments are tested exactly for the exponential distribution at the 0.00001 significance level, i.e. one thousandth of one percent. When we form a synthetic distribution, only the top of this distribution will be blurred as small deviations correspond more to market noise and it is here where the instruments have the maximum degree of independence between each other which allows us to apply the DPT and assert that in a limited area near zero the synthetic distribution will be distributed close to the normal law and the greater the number of initial FI, the closer it is. But on the tails the situation is opposite - broad market movements are very strongly correlated between different FI, as they are caused, as a rule, by fundamental reasons. As a consequence, in the area of large deviations the independence condition is not fulfilled and the TPT does not work: the exponential tails remain as they were.
So if you try to identify the synthetic distribution as any of the known ones, it probably won't work, as it is rather some mix of two different laws.
Nevertheless, my request to hrenfx stands - to give an example of a synthetic instrument that is supposed to have significantly different characteristics from individual FIs. Purely for research purposes, just to see if the performance improvement actually exists or is a sham.
It's not a problem. Just pick up the dynamic coefficients for the instruments that make up the synthetic, and you can generate a BP of any shape. The problem is that the word "fit" means to fit and surrenders the synthetic BP graphs presented by hrenfx with just such dynamic coefficients. But if he was able to cross, say, WTI with CRB without any dynamic coefficients and get a fundamentally different picture of the same returns or autocorrelations - that would be great.
Perhaps, a greater interest represents neutral CVs, the totality of distributions of which significantly differs from the normal one, but apparently this is the topic of another branch.
If it were that easy everyone would trade for example seasonal spreads and be happy.
But if he could cross, say, WTI with CRB without any dynamic coefficients and get a fundamentally different picture of the same returns or autocorrelations - that would be great.
Above was an example of a stationary synthetic.
In the general case, stationarity is not needed because the coefficients are dynamic.
Now it is clear that it is calculated by an absolutely meaningless formula:
USDLFX = ((USDJPY * USDCHF * USDCAD) / (EURUSD * GBPUSD * AUDUSD))^(1 / 7)
P.S. Even this version of the index makes more sense (there would be degree = 1 / 6).
The seventh degree root is correct. The sixth is not.
The matter is that in implicit calculation of the index all the factors are multiplied by one = USD/USD
i.e. the full formula is: USDLFX = ((USDUSD *USDJPY * USDCHF * USDCAD) / (EURUSD * GBPUSD * AUDUSD))^(1 / 7)
And that is correct.
can anything be "squeezed" out of the bicurrency basket in terms of arbitrage?
although I suspect there is nothing in the forecasting part either (((.
Тем не менее, моя просьба к hrenfx в силе - дать пример синтетического инструмента, который, как предполагается имеет существенно отличные характеристики от отдельных ФИ. Чисто в исследовательских целях, просто чтобы убедиться, существует ли улучшение характеристик на самом деле или это фикция.
The request still stands. We are waiting for the topicstarter's convincing examples.