3
$\begingroup$

I have a set of "Days since last Visit" data from the Google Analytics:

Day since last Visit     Count of Users
Total                    400K
0                        100K
1                        50K
2                        10K
....

Given a fixed Time Frame, the row "1 50K" means that there are 50K users who have the last two visits in a consecutive day, and "0 100K" means that for that 100K users, their last two visits occur on the same day.

So is it possible for me to estimate, the expected number of days for the users visit the sites in a month?

$\endgroup$
2
  • $\begingroup$ Do you mean you want to estimate the expected number of visits per month of a user; or do you just want the expected value of the "days since last visit" variable? $\endgroup$ – Peter Ellis Jun 25 '13 at 5:32
  • $\begingroup$ expected Days of visits per month of a user, so if the user come several times in the same day, that only count as 1. $\endgroup$ – Kloser Cheung Jun 25 '13 at 11:27
1
$\begingroup$

It is valid to find the expected number of days between visits and then divide the number of days in a month by the expected days between visits to get the vitis per month

$\endgroup$
3
  • 1
    $\begingroup$ Yes, I am now doing something similar. I am just afraid that is not an accurate estimation. I was thinking whether this can be transformed into a Poisson Question. But I dont know whether I can do so, and how I can do so. $\endgroup$ – Kloser Cheung Jun 25 '13 at 11:29
  • $\begingroup$ The problem will be that this counts visits, not days that include a visit. So it will give too large an estimate. $\endgroup$ – Peter Ellis Jun 25 '13 at 21:14
  • $\begingroup$ That is exactly the problem i am facing $\endgroup$ – Kloser Cheung Jun 26 '13 at 3:11

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.