On the unequal probability of a price move up or down - page 159
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
There is no cointegration with vector (1, -1), but there may (or may not) be cointegration with another vector of cointegration.
Yes, if you dynamically manipulate (after an unsuccessful entry) total lots in each leg, taking into account changing situation, somewhere to add, somewhere to subtract, then you can achieve profit. But can such trading based on player's skill be called pair trading? In the classical case it is based on the principle of "open and forget" - the profit is guaranteed,
I think 2X-Y should be good. Although the k's do change over time, of course.
This is all true, but provided we know in advance the future behaviour of the pairs. What if pair X stops, will its doubled lot help then?
After all, in the classic version it assumes the principle of open and forget - profit is guaranteed. ,
It's more of a mythical option. But everyone is chasing it for some reason.
Maybe instead of guessing on a triangle, we could use real statistics?
For example, for the pound to take the history of quotes, report like the attached: somehow convert the turnover from USD to GBP, then on the hypothesis that the total number of pounds per month does not change much count the weighting coefficients and identify the overbought/oversold.
Yes, if you dynamically manipulate (after an unsuccessful entry) total lots in each leg, taking into account changing situation, somewhere to add, somewhere to subtract, you can achieve profit. But can such trading based on player's skill be called pair trading? In the classical case it is based on the principle of "open and forget" - the profit is guaranteed,
And what if, in addition to the non-stationarity that cointegration can help to solve, there can be other types of cointegration? This is for example what in econometrics is called structural jumps.
This is all true, but provided we know in advance the future behaviour of the pairs. What if pair X stops, would doubling its lot help then?
Of course it won't, but the highest bidder will.
Personally, I only have one question.
which way is the market going - against higher volume or in the direction of increasing margin in the long term?
It's more of a mythical option. But everyone is chasing it for some reason.
It's a feasible option, if you can find a good cointegrated pair. But will you find any? It is possible, of course, to create synthetic pairs with such property, but again there is a problem: how to provide stability of cointegration.
I personally have only one question.
Is the market moving against a higher volume or towards a rising margin?
There is no such correlation.
If there were an unambiguous answer to such simple questions, then the ratio of successful traders to losers would be inverse to the existing one.
By the way, what is the "growing margin" and how do you define it?
Of course it won't, but it'll be the bigger pair that haggles
I personally have only one question
Where does the market go - against higher volume or in the direction of margin increase in the future?
the market always moves towards the change of the current trend ;)
Yes, TC has given a lot of attention to correlation, but not a word about cointegration.
Co-integration in forex is far from reality.