Statistical independence and model choice: An example (Q1124697)

The author considers the following lottery situation problem: after \(n\) consecutive playings at the lottery and not winning, is the probability of winning at the \((n+1)\) playing increased or not? The aim of this note is to present an analysis of this classical lottery problem by using two probabilistic models, accessible to any undergraduate student in mathematical statistics. The frequentist and Bayesian models differ mainly with respect to the independence of the involved events. However, not surprisingly, the overall conclusion from either model is the same: it is very difficult to win at the lottery!