relative drift of two stocks

[Gazillionaire]

Cointegration is much harder to define in laymans terms without being inaccurate. In slightly more complex terms, if the price series x and y are cointegrated, then there exists some equation y_t = a + bx_t + e_t where the error term e is stationary i.e. it's not drifting and has constant expected volatility.
Rob

What purpose does this serve?
In pairs trading, we want to find some parameter which governs the statistical tendency for two price lines to intersect on an overlay chart or a single price ratio line to revert to the mean of a moving average.

 

[globalarbtrader]

What purpose does this serve?
In pairs trading, we want to find the parameter/statistical tendency for two price lines to intersect on an overlay chart or a single price ratio line to revert to mean moving average.

They are both actually special cases of the more general case.

If the lines overlap, then b=1
If a ratio line reverts, then b=ratio

Rob

 

[2rosy]

I noticed that in many cases, even when the correlation % is very high, on average one of the two tends to go up more and go down less, and so over time there will be a noticeable difference on the returns.

How are you determining this? absolute values, pct chg, log diff, weightings?

 

[earth_imperator]

I'm recently playing a lot with correlation trades..

I noticed that in many cases, even when the correlation % is very high, on average one of the two tends to go up more and go down less, and so over time there will be a noticeable difference on the returns. For lack of better terms, I'll call this "relative drift"

Would it be useful for trading if there existed 2 that are the exact inverses of each other?
What about BOIL and KOOL ?

Is there any mathematically sound/consolidated practice to measure this factor?

What about the deviation rate (for example %) from a reference r in time?

 

[earth_imperator]

Would it be useful for trading if there existed 2 that are the exact inverses of each other?
What about BOIL and KOOL ?

Sorry, BOIL and KOLD was meant. Both are inverse to each other. They are ETFs.

What about the deviation rate (for example %) from a reference r in time?

 

[Baozi]

How are you determining this? absolute values, pct chg, log diff, weightings?

normally percent change.