Econometrics: why co-integration is needed - page 13
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
At the same time, at the right point, we buy the euro and sell the inverse dollar index, like a loca. Then at another point we close both operations and open reverse ones, i.e. we sell the euro and buy the inverse index. Here the whole point is in the rate of return per unit time - judging by your pictures roughly it is something like 30-40 pips per week. If we exclude spread/swap in instruments and slippage at opening/closing we get 15-20 pips per week. In general it is not bad as long as it is stable and this is what we should check very attentively.
San Sanych, how about this?
I'll give it a try.
San Sanych, can't you do it this way?
Here's the result. Cotier and balance graphs
Interestingly, they are not similar.
Recording the cointegration regression
Evaluation of cointegration vector
The vector is decent enough. If we divide 100% by the values in t_Statistic column, we will get the error of evaluation of the appropriate vector coefficient in %. The values below are not very noteworthy, although the R-square is very nice.
The result of cointegration in graphical form:
You can clearly see that kotig and balans are cointegrated.
I don't know about the collective, but I would conclude:
Since the difference between kotig and balans is stationary, you can trust the tester to test the TS. How long will such happiness last? I don't know. Since this result is based on the fact that both I(1) series are integrated.
For 15-20 pips per week we should not discard the swap/spread. Hopelessness and stability have nothing to do with it.
It can be used as a hedge in combination with a breakout system. There is always an open synthetic pair and if at a certain moment the breakdown shows that you should open a buy position in e.g. euro, then the corresponding instrument will be closed.
I was only saying about 40 pips, we should test everything there.
It can be used as a hedge in combination with a breakout system. There is always an open synthetic pair and if at some moment the breakdown shows that we should open a buy position in e.g. euro, we simply close the corresponding instrument.
And about 40 pips is just a hunch, we should test everything there.
The idea of cointegration is to find a statistically guaranteed coordinated movement of the instruments. We would be interested in a variant of the guaranteed inconsistent movements of the instruments, that is called negative correlation, but with the difference as a stationary process.
Probably so.
With cointegration, you can see well from the picture. Take TC: entry at zero balance chart crossing and exit. The income is zero (excluding overheads). But there is a maximum between these two points. If we start to catch it, we get all the problems of a regular TS.
If we can prove that profitable TPs are cointegrated with cotier and unprofitable TPs are not cointegrated, it will revolutionise the testing of TPs
Send a .csv cotri balance. Preferably both profitable and unprofitable TS. Any. Ready to do the calculations and post the result.
faa1947: Попробовал взять на чемпионате, но не получилось скопировать. Кто-нибудь знает как там скопировать?
Right-click on the equity graph - "Export that ...."
It doesn't work. And I don't need a graph, I need a table. The first page is copied, the subsequent ones are not.
If it can be proven that profitable TS are cointegrated with a quotient and unprofitable TS are not cointegrated, then that is a revolution in TS testing
Send the .csv cotri balance. Preferably both profitable and unprofitable TS. Any. Ready to do the calculations and post the result.
Thank you for the interesting result and the crystal clarity of the way it is achieved.
I'm sure the result is intermediate.