I have a dataset of open, high, low and close values for several financial instruments. I wish to determine if any of these instruments have statistically significant values on specific days. I would then like to determine which of these instruments produce statistically significant values on the same days.
My thinking is that I first need to know what the distribution of the data is to know what is 1 and 2 SD away from the median.
But how do I testing for how consistently these instruments produce values beyond 1 or 2 SD on specific days?
How do I determine which of the instruments produce correlated statistically significant values on the same days?