2 ms·
Those are good points. The event in question is not a simple user visit, however. Let's say it's a particular user action that's a bit more involved, e.g. filli
by nicolewhite 11y ago
Those are good points. The event in question is not a simple user visit, however. Let's say it's a particular user action that's a bit more involved, e.g. filling out a form. The assumption here is that users fill out this form independently from one another.
- mturmon 11y agoAh, I just noticed you are the OP. Sure, you have to start somewhere with any analysis, and Poisson is the place to start. Anyone (like me) can then question independence assumptions from the sidelines. I was moved to speak up originally because a commenter referred to inspecting the logs to verify the Poisson assumption, and in my experience, looking at the data is always an excellent principle, and generally preferable to just stopping at an intuitive understanding of the arrivals process.