Beruflich Dokumente
Kultur Dokumente
𝑓 𝑥 = 𝑃 𝑋 = 𝑥 = 𝑝𝑞 𝑥−1
https://onlinecourses.science.psu.edu/stat414/node/75
Geometric Distribution Proof
•
=> f(x) = p*(1 – p)(x – 1), where, x = 1, 2, 3, 4, . . . . . .
Now, let us plug in different values for ‘x’ and determine the probability in each
case.
f(1) = p * q(1 – 1) = p, f(2) = p * q(2 – 1) = p * q, f(3) = p * q(3 – 1) = p * q2 and so
on for higher values of x.
Since we do not know when the first success occurs, the value of ‘x’ might
extend up to infinity. Thus, the probability of the variable X is a sequence like
p, p(q), p(q)2,. . . . . . . It is very clear that the sequence is a geometric sequence
and hence the probability function P(X) can be defined as geometric function.
• http://math.tutorvista.com/statistics/geometric-distribution.html
Mean of Geometric Distribution
• The geometric distribution is the appropriate distribution when the probability is obtained of the
single success occurring on the x-th trial. Geometric probability distribution is a statistical data. In
any statistical data it is important to know the mean and variance.
• We know, f(x) = p * q(x – 1), where ‘x’ takes the values from 1 to infinity.
• Now let us check what the infinite sum ∑f(X) is.
∑f(X) = p + pq + p(q)2 + . . . . . . = p(1 + q + q2 + . . . . .)
• The series inside the parenthesis is an infinite geometric series with a common ratio of ‘q’.
∑f(X) = p + pq + p(q)2 + . . . . . . = p[1/(1 – q)] = p/p = 1. In other words the geometric
series is convergent and converges to 1.
• The mean of the geometric distribution µ is defined as, ∑x * p * q(x – 1)
So, µ = ∑xf(X) = p + 2pq + 3pq2 + . . . . . .
Multiplying both sides by ‘q’, µq = pq + 2pq2 + 3pq3 + . . . . . .
Now, µ - µq = [ p + 2pq + 3pq2 + . . . . .] – [pq + 2pq2 + 3pq3 + . . . . . .]
Or, µ(1 – q) = p + pq + p(q)2 + . . . . . . = p[1/(1 – q)] = p/p = 1.
1 1
E X =𝜇= =
1−𝑞 𝑝
Expected Value of Geometric
Distribution
• In a probability distribution the expected value is towards the
mean if the distribution is convergent. In case of geometric
probability distribution we had seen it is a geometric function
and converging to 1. Therefore, for a geometric probability
distribution the expected value is same as the mean. In notation,
the expected value is denoted as E(X).
1
• 𝐸 𝑋 =𝑝
• By using same method of proofing we have
𝑞
𝑉𝑎𝑟 𝑥 =
𝑝2
Example 1(a)
• League’s Marketing Division randomly selects people on a random street in
Kansas City, the last home football game. Let p, the probability that he
succeeds in finding such a person, equal 0.20. And let X denote the number of
people he selects until he finds his marketing representative must select 4
people before he finds one who attended the last home football game?
• Solution:
To find the desired probability, we need to find 𝑃 𝑋 = 4 , which can be
determined readily using probability mass function of a geometric random
variable with 𝑝 = 0.20, 𝑞 = 0.8 𝑎𝑛𝑑 𝑥 = 4.
• Solutions:
To find the desired probability we need to find 𝑃 𝑋 > 6 , which
can be determined readily using cumulative distribution function of
a geometric random variable with q = 0.8 𝑎𝑛𝑑 𝑥 = 6