Prior Probability

A prior probability is a probability based on data from the past, or the probability of an event occurring before new data has been added to the model.

Let’s say we’re on Let’s Make a Deal. We’re shown three doors. The NEW CAR!!! could be behind any of the three doors, so the prior probability of us picking the NEW CAR!!! door is one in three, or 33%. When Wayne opens a door not hiding the NEW CAR!!!, the posterior (or “after”) probability of us picking the correct door has upgraded to one in two, or 50%. Right? We've gotten rid of one possibility, so now only 2 exist. We can completely forget about the zonk behind Door Number 3.