|
|
|
Here's what you'll find in this section:
Some Discrete Distributions
Perhaps the most commonly used discrete probability distribution is the binomial distribution. An experiment which follows a binomial distribution will satisfy the following requirements (think of repeatedly flipping a coin as you read these):
, is constant from one trial to another.The random variable X of a binomial distribution counts the number of successes in n trials. The probability that X is a certain value x is given by the formula
where 
Recall that the quantity 
, ``n choose x,'' above is
where
We could use the formulas previously given to compute the mean and variance of X. However, for the binomial distribution these will always be equal to
NOTE:\ A random variable X
which follows the binomial distribution can be abbreviated 
, where 
is read ``is distributed as.''
NOTE:\ A particularly important
example of the use of the binomial distribution is when sampling with
replacement (this implies that 
is constant).
EXAMPLE:\ Suppose we have 10 balls
in a bowl, 3 of the balls are red and 7 of them are blue. Define success
S
as drawing a red ball. If we sample with replacement, P(S)=0.3
for every trial. Let's say n=20, then 
and we can figure out any probability we want. For example,
The mean and variance are
EXAMPLE:\ Assume we are playing a
card game with a regular deck of 52 cards, where 16 of these are ``face
cards'' and each ``hand'' consists of 10 randomly selected cards. Using
combinations, find the probability of getting 4 face cards in a hand of 10
cards. We know there are 
possible ``hands'' and 
possible hands with 4 face cards, therefore
The distribution we derived above is called the hypergeometric distribution. If we can define objects as success/failure, and we have N total items (N=52 cards), with M successes in the population (M=16 face cards) and n items chosen (n=10 cards per hand), then
For our previous example,
For the hypergeometric distribution,
If we let 
, then
Note that except for 
, this is the same as the mean and the variance for the binomial
distribution. C is called the finite population correction.
The negative binomial distribution is used when the number of successes is fixed and we're interested in the number of failures before reaching the fixed number of successes. An experiment which follows a negative binomial distribution will satisfy the following requirements:
, is constant from one trial to another.
. Its probabilities are computed with the formula
for 
Formulas for E(X) and 
for the negative binomial distribution are given by
EXAMPLE:\ Suppose we are at a rifle
range with an old gun that misfires 5 out of 6 times. Define ``success'' as
the event the gun fires and let X be the number of failures before
the third success. Then 
. The probability that there are 10 failures before the third success is
given by
The expected value and variance of X are
The Poisson distribution is most commonly used to model the number of random occurrences of some phenomenon in a specified unit of space or time. For example,
where 
is the average number of occurrences in the specified interval. For the
Poisson distribution,
EXAMPLE:\ The number of false fire alarms in a suburb of Houston averages 2.1 per day. Assuming that a Poisson distribution is appropriate, the probability that 4 false alarms will occur on a given day is given by
Applicable StataQuest Commands:
Statistics 
Correlation
Pearson (or Spearman)
Statistics
Simple regression to obtain the coefficients for the least square
(regression) line
Graphs
Scatterplots
Plot Y vs. X, with regression line
%
The webmaster and author of this Math Help site is Graeme McRae.