Econometrics: why co-integration is needed - page 4
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
The correlation of the two time series does not correlate with the profit of the TS ))))
That's for sure. I've tried modelling the profit as a function of the quote and it didn't work. Although it is suggestive. After all, there is a TS between the quote and the profit, i.e. some function that transforms the quote into a profit....
San Sanych, is it already on the subject of "Tyapnitsa"? :)
I wonder if kotir and profit are cointegrated or not?
If they are not cointegrated, then the profit is random. And if they are cointegrated, then the profit is not random?
If someone gives me a file with two rows: kotir and profit, then I will calculate the cointegration. Very curious.
If they are not cointegrated, the profits are incidental.
Only if you identify randomness with non-linearity :)
Faa, explain something to me on your fingers. Well, let's say a false correlation.
I don't get it. I'll read what false correlations are.
Very simple. We take two series and calculate the correlation using a formula. You always get a number and never no number. I.e. the calculation always gives a correlation value between anything. The big experts in this field are astrologers.
Nothing else comes to mind.
Although.
The alpha and omega of statistics is the need to match the economic (physical) meaning of any figure. If we could, it means that the figure we got from the correlation formula has a meaning: -1, 0 or +1 or whatever. If we cannot correlate Saturn's rings (perhaps we cannot yet), then any calculation of the correlation of Saturn's rings with the cotier is a false correlation.
Only if you identify randomness with non-linearity :)
I don't really get it. To me, non-linearity is a deterministic process.
Co-integration is a very powerful concept. Two random processes are considered co-integrated if the difference between them is a stationary random process. Trends and biases, if present in random processes, must be removed.