Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

18 Statistics for mining engineering

Figure 1.11. The Bernoulli probability density function.

Geometric distribution
This distribution is defined in a twofold manner:
a. as the probability distribution of the number X of independent Bernoulli trials needed to
get one success; this distribution is supported on the set {1, 2, …}
b. as the probability distribution of the number Y = X – 1 of failures before the first success;
this distribution is supported on the set {0, 1, 2, …}
If the probability of success in every trial is p, then the probability that in k-th trial will be
success is determined by the pattern:

P(X = k) = (1 – p)k−1p for k = 1, 2, … (1.41a)

Similarly, if the probability of success in every trial is p, then the probability that there will
be k-th failures preceding the first success is determined by the pattern:

P(Y = k) = (1 – p)kp for k = 0, 1, 2, … (1.41b)

Often, the name shifted geometric distribution is adopted for the distribution of the number
X. This ‘displacement’ is visible looking at the mode because for X the random variable is 1
whereas for the second random variable it equals 0.
The expected value and the variance in these probability distributions are the following:

E(X) = 1/p E(Y) = (1 – p)/p σ2(X) = σ2(Y) = (1 – p)/p2 (1.42)

The geometric probability distribution is the discrete equivalent of the exponential distri-
bution18 and it has no memory as does its exponential counterpart.19
An example of the probability mass distribution for geometric distribution is given in
Figure 1.12.
Considering the number of hoist winds up to the moment of the over-winding at an extreme
level, it is easy to see that this number is a random variable that is geometrically distributed in
which p denotes the probability of over-winding in a single hoist work cycle.

18
This distribution is given by formula (1.52).
19
The idea of a lack of memory is extended into the area of stochastic processes, e.g. the Poisson process
in which the time from the occurrence of one event to the next one is an exponential one that is charac-
terised only by lack of memory. This lack of memory means that the probability of the appearance of
k events in the interval (T, T + t) does not depend on how many events and in which manner they have
occurred up to the moment T (Gnyedenko and Kovalenko 1966).

Book.indb 18 12/9/2013 12:22:08 PM

You might also like