5
$\begingroup$

I have two time-series:

  1. Number of emails opened daily between January-March.
  2. Number of applications created daily between January-March.

There is an expected time lag between someone opening an email and actually creating an application, it does not have to happen the same day. How can I account for this and test whether number of emails opened affects the number of applications created?

Plotting a scatter plot in Python does not show much as each point is strictly the data from the same date.

$\endgroup$
1
  • 2
    $\begingroup$ Have you tried plotting the two series, while sequentially lagging one of the series, to see if and when they match? $\endgroup$ – user2974951 Aug 29 '19 at 10:09
3
$\begingroup$

If there is a time lag, the first thing I can think of is to estimate a VAR model and use a Wald test to test the significance of the coefficients of the lagged values of emails opened in the equation for the current value of the number applications created (dependent variable). See pag. 388 of this source on VAR and Wald

$\endgroup$

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.