I know its not a good idea doing mean reversion strategies on stocks but i just wanted to test if AAPL and AMZN revert away from each other and then come back to the mean.
In the notebook i attached, i checked for a lagging correlation ( i shifted AMZN's stock price one place up. Left apple alone. --- ( Followed quantopians useful advice / lectures for all of this) --- )
By looking at any daily chart.. It does seem like there is a divergence/ lagging correlation. ( I just wanted to test and get more evidence.)
Is there any way to detect if there were periods of time when AAPL and AMZN diverged by a significant amount? Maybe by using other statistical methods -- or anything just other method?
The lagged correlation is definitely not enough evidence. Its shifted one date/price up --- so the price change didn't happen at the same time.