Geometric distribution:
(1) the number of something member (also called test number) are fixed, it is represented by n. (E.g., a coin toss three times, 5 times a program execution)
(2) Each event has two possible outcomes (success or failure). (Eg, program execution (success), program execution (failed))
(3) Each time the probability of "success" is equal to the probability of success is represented by p.
(4) It is also and that is where the difference between binomial, binomial probability of success of solving the problem is x times. The geometric distribution problem solving becomes a - test x times before they get the probability of success for the first time. For example, the implementation of 101 times, 101 times before it can be executed correctly (the result is not procedural, the other software conflicts) probability.
Where, p is the probability of success, x represents the number of trials.
Assumed probability of success is 0.2, the probability of success of execution 101 is: p (101 execution successful) = (1-0.2) * 0.2 = 100 ^ 100 ^ 0.8 0.2 *
Is desired geometric distribution E (x) = 1 / p, the minimum number of each successive
Geometric standard deviation of the distribution:
Binomial distribution:
(1) the number of something member (also called test number) are fixed, it is represented by n. (E.g., a coin toss three times, 5 times a program execution)
(2) Each event has two possible outcomes (success or failure). (Eg, program execution (success), program execution (failed))
(3) Each time the probability of "success" is equal to the probability of success is represented by p.
Probability (4) the success of x is the number of times, for example, to perform 101 times, 99 times can perform the correct probability.
Where n represents the number of incidents, x represents the number of successes, p is the probability of success every time, p (x) is performed n times, there are x times the probability of success
Expected: E (X) = np
Poisson distribution:
(1) events are independent events
The same probability (2) at any same time, events (e.g., within a probability of winning, the winning probability is the same as on day 2)
(3) within a certain range, the probability of occurrence of x is something much (for example, a shopping mall staged a promotional sweepstakes, want to know the probability of winning five people in one day)
x is the number of successes, u is the average number of occurrences in the interval
Special is its expectation and variance are equal to u
Normal distribution: Click the link