> shouldn't be constant and is heavily dependent on time/date. A priori it is assumed that sales are independent of time. That's part of what Poisson distribu

> shouldn't be constant and is heavily dependent on time/date. A priori it is assumed that sales are independent of time. That's part of what Poisson distribution means -- a constant rate of rolls of a fixed weighted die. The assumption could be wrong. > justification for why sales mean should equal sales variance https://proofwiki.org/wiki/Variance_of_Poisson_Distribution https://proofwiki.org/wiki/Variance_of_Poisson_Distribution