... Unless your application meets these assumptions, treat the results of the model with caution. Models seldom match real processes perfectly, but we need for them to be close. If not, the results can be misleading.
Finite Populations
One reason for the mismatch between Bernoulli trials and situations encountered in practice is the finite size of real populations. In the example of detailing, suppose that 40% of all receptionists in the territory are willing to let the detail rep speak to the doctor. Once the rep succeeds, there’s one less willing receptionist for the next visit. Each success makes it more likely that the next receptionist blocks the path. The trials are consequently dependent.
A familiar illustration of this type of dependence occurs ...
Get Statistics for Business: Decision Making and Analysis, 3rd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
