0
$\begingroup$

I am doing correlation test on a factor and stock price(both are numerical values). I calculate the correlation from the following two ways:

  1. Pearson correlation between factor value and stock price
  2. Pearson correlation between factor value and first order difference of stock price(i.e., price[i+1] - price[i])

It turns out correlations in two cases vary a lot. Any explanation on the difference between the two approaches? I used to suppose they are very close.

Improve this question
$\endgroup$
1
$\begingroup$

Price and it's first order difference are not very close (maybe they're even independent!), so their correlations with any other variable do not have to be close. Toy example:

factor   price[i]   price[i+1]   difference
     1          1            3            2
     2          2            3            1
     3          3            3            0

Then correlation between factor value and stock price is 1 and correlation between factor value and first order difference of stock price is -1.

Improve this answer
$\endgroup$
  • $\begingroup$ yes, they price and price difference are not necessarily correlated. $\endgroup$ – franky Jul 4 '17 at 8:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.