1, random experiment
Features: All possible results can be listed by it. (In endless case, at least theoretically)
2, probability definition
Relative frequency definition: p (e1) = LIM (N1 / N)
P (e1) = 1, E1 is an inevitable event;
P (E0) = 0, E1 is an impossible event.
3, incompatible event: p (a or b) = p (a) p (b) (Note: "a or b" means A occurrence or B occurs)
Index: p (a and b) = p (a) p (b) (Note: "A and B" means A occurrence, B also happens)
4, Bayesian theorem
(On school)
Time to listen
The teacher said, did not understand, this time it is readily understood)
The Bayesian method is actually a deductive method (by general to special, also known as the Holmes law). It is known for a system event B, and ask for a certain reason event AK's probability. For example, the Titanic is known to sink, and the probability of imposing the problem of instrument.
An event B causes the reason for event B as a series of incompatible events {A1, A2, ..., AK, ...} = a. The probability of each cause event (P (AK)) is known, the reason for the difference (P (B | AK)) is known, then in the case of B, the probability caused by one reason (P (AK | B))
5, two distributions
One of two possible items per experiment.
6, probability density
7, accumulate distribution functions
