Binomial Random Variables

Suppose that n independent sub-experiments (or trials) are

performed.

Each of which results in either a “success” with probability p

Or a “failure” with probability 1 − p.

If X is the total number of successes that occur in n trials,

Then X is said to be a binomial random variable with parameters

n and p.
 Binomial Random Variables
Suppose that n = 3 and that we are interested in the probability that X is
equal to 2.

That is, we are interested in the probability that 3 independent trials,

each of which is a success with probability p, will result in a total of 2
successes. The sample space for X =2 success will be

(s, s, f ), (s, f , s), (f , s, s)

P(s, f, s) = P(S1F2S3) = P(S1)P(F2)P(S3)

= p(1-p)p By Independence
 Binomial Random Variables

Since each of the 3 outcomes that result in a total of 2 successes

consists of 2 successes and 1 failure,

it follows in a similar fashion that each occurs with probability

p2(1 − p).

Therefore, the probability of a total of 2 successes in the 3 trials is

3p2(1 − p).
 Binomial Random Variables
Consider now the general case in which we have n independent trials. Let X denote the
number of successes, then
P{X = i} = total of i successes.

Since this outcome will have a total of i successes and (n − i) failures, it follows from the
independence of the trials that its probability will be

pi(1 − p)n−i

Total number of outcomes with exactly i successes and (n − i) failures for n events is

n!/[i!(n − i)!]
 Binomial Random Variables
The probabilities of three binomial random variables with respective parameters
n = 10, p = 0.5, n = 10, p = 0.3, and n = 10, p = 0.6 are
 Binomial Random Variables
Three fair coins are flipped. If the outcomes are independent, determine the probability
that there are a total of i heads, for i = 0, 1, 2, 3.
 Binomial Random Variables
Suppose that a particular trait (such as eye color or handedness) is determined by a single pair
of genes, and suppose that d represents a dominant gene and r a recessive gene. A person with
the pair of genes (d, d) is said to be pure dominant, one with the pair (r, r) is said to be pure
recessive, and one with the pair (d, r) is said to be hybrid. The pure dominant and the hybrid
are alike in appearance. Their offspring receives one gene from each parent, and this gene is
equally likely to be either of the parent’s two genes.

(a) What is the probability that the offspring of two hybrid parents has the opposite (recessive)
appearance?
(b) Suppose two hybrid parents have 4 offsprings. What is the probability 1 of the 4 offspring
has the recessive appearance?

(a) The offspring will have the recessive appearance if it receives a recessive gene from
each parent. By independence, the probability of this is (1/2)(1/2) = 1/4.
(b) Assuming the genes obtained by the different offspring are independent (which is the
common assumption in genetics), it follows from part (a) that the number of offspring having
the recessive appearance is a binomial random variable with parameters n = 4 and p = 1/4.
Therefore, if X is the number of offspring that have the recessive appearance, then

= 27/64
 Binomial Random Variables
Suppose that X is a binomial random variable with parameters n and p, and suppose we want
to calculate the probability that X is less than or equal to some value j. In principle, we could
compute this as follows:
 Binomial Random Variables
Expected Value and Variance of a Binomial Random Variable
A binomial (n, p) random variable X is equal to the number of successes in n independent
trials when each trial is a success with probability p.

Therefore, using the fact that the expectation of the sum of random variables is equal to the
sum of their expectations, we see that

E[X] = np
Also, since the variance of the sum of independent random variables is equal to the sum of
their variances, we have

Var(X) = np(1 − p)
 Binomial Random Variables
A communications channel transmits the digits 0 and 1. Because of static, each digit transmitted
is independently incorrectly received with probability 0.1. Suppose an important single-digit
message is to be transmitted. To reduce the chance of error, the string of digits 0 0 0 0 0 is to be
transmitted if the message is 0 and the string 1 1 1 1 1 is to be transmitted if the message is 1. The
receiver of the message uses “majority rule” to decode; that is, she decodes the message as 0 if
there are at least 3 zeros in the message received and as 1 otherwise.
(a) For the message to be incorrectly decoded, how many of the 5 digits received would have to
be incorrect?
(b) What is the probability that the message is incorrectly decoded?

a) Three

b) Assume random variable is the transmission of incorrect signal with p of 0.1 and n =5

Now calculate the probability for X =3, X=4 and X=5 and then add them for the answer

10
 Hypergeometeric Random Variables

Suppose that n batteries are to be randomly selected from a bin of N

batteries

Out of which Np are functional and the other N(1 − p) are defective.

The random variable X, equal to the number of functional batteries

in the sample is then said to be a

Hypergeometric random variable with parameters n, N, p.

 Hypergeometeric Random Variables

Hypergeometric random variable with parameters n, N, p

Number of Trial = n , total size N and

probability of successful trial i = p

What distinguishes X from a binomial random variable is that these

trials are not independent.
 Hypergeometeric Random Variables
For instance, suppose that two batteries are to be withdrawn from a bin
of five. Only one is functional and the others defective.
(That is, n = 2, N = 5, p = 1/5.)
Then the probability that the second battery withdrawn is functional is
= 1/5
However, if the first one withdrawn is functional, then the conditional probability that
the second one is functional is 0 (since when the second battery is chosen all four
remaining batteries in the bin are defective).

That is, when the selections of the batteries are made without replacing
the previously chosen ones, the trials are not independent, so X is not a
binomial random variable.
 Hypergeometeric Random Variables

By using the result that each of the n trials is a success with probability p, it can
be shown that the expected number of successes is np. That is,

E[X] = np

N - n np (1-p)
Var (X) =
N-1

Thus, whereas the expected value of the hypergeometric random

variable with parameters n, N, p is the same as that of the binomial
random variable with parameters n, p, its variance is smaller than
that of the binomial by the factor (N − n)/(N − 1).
 Hypergeometeric Random Variables
If 6 people are randomly selected from a group consisting of 12 men
and 8 women, then the number of women chosen is a hypergeometric
random variable with parameters n = 6, N = 20, p = 8/20 = 0.4.
Its mean and variance are

E[X] = np = 6(0.4) = 2.4

N - n np (1-p) = (14/19)6(0.4)(0.6) ~ 1.061

Var (X) =
N-1

Similarly, the number of men chosen is a hypergeometric random

variable with parameters n = 6, N = 20, p = 0.6
 Hypergeometeric Random Variables
suppose that 20 batteries are to be randomly chosen from a bin containing 10,000
batteries of which 90 percent are functional. In this case, no matter which batteries
were previously chosen each new selection will be defective with a probability that
is approximately equal to 0.9. For instance, the first battery selected will be
functional with probability 0.9. If the first battery is functional then the next one
will also be functional with probability 8999/9999 ≈ .89999, whereas if the first
battery is defective then the second one will be functional with probability
9000/9999 ≈ .90009.

A similar argument holds for the other selections, and thus we may conclude that
when N is large in relation to n, then the n trials are approximately independent,
which means that X is approximately a binomial random variable.

Problem Set on P 249

 Poisson Random Variables
Consider Binomial random variables with n independent trials,
each of which results in either a success with probability p
or a failure with probability 1 − p.

If the number of trials is large and the probability of a success on a trial

is small, then the total number of successes will be approximately a
Poisson random variable with parameter λ = np.

Examples:
1. The number of misprints on a page of a book
2. The number of people in a community who are at least 100 years old
3. The number of people entering a post-office on a given day
 Poisson Random Variables
A Random Variable X that takes one of the values of 0, 1, 2,.. is said to
be a POSSION RANDOM VARIABLE with parameter l,
if for some positive value of l, its probabilities are given by

P{X = i} = (c λi) / i! , i = 0, 1, ...

Where c is constant and its value depend upon λ
c = e−λ
P{X = i} = (e−λ λi) / i! , i = 0, 1, ...

If X is POSSION RANDOM VARIABLE with parameter l, l > 0, then

E[X] = l and Var(X) = l

 Poisson Random Variables
Probabilities of a Poisson random variable with λ = 4.
 Poisson Random Variables
If X is a Poisson random variable with parameter λ = 2, find P{X = 0}.

P{X = 0} = (e−2 20) / 0!

Now 20= 1 and 0! = 1, therefor

P{X = 0} = e−2 = 0.1353

 Poisson Random Variables
Suppose that items produced by a certain machine are independently defective with
probability 0.1. What is the probability that a sample of 10 items will contain at most 1
defective item? What is the Poisson approximation for this probability?

If we let X denote the number of defective items, then X is a binomial

random variable with parameters n = 10, p = 0.1.
Thus the desired probability is:

P{X = 0} + P{X = 1} = 10C0 (0.1)0(0.9)10 + 10C1 (0.1)1(0.9)9

= 0.7361
Since np = 10(0.1) = 1, the Poisson approximation yields the value
P{X=0}+P{X=1}= e−1+e−1 = 0.7358

Thus, even in this case, where n is equal to 10 (which is not that large)
and p is equal to 0.1 (which is not that small), the Poisson approximation
to the binomial probability is quite accurate.
 Poisson Random Variables
Suppose the average number of accidents occurring weekly on a particular highway is
equal to 1.2. Approximate the probability that there is at least one accident this week.

Let X denote the number of accidents. Because it is reasonable to suppose that there are
a large number of cars passing along the highway, each having a small probability of
being involved in an accident, the number of such accidents should be approximately a
Poisson random variable.

That is, if X denotes the number of accidents that will occur this week, then X is
approximately a Poisson random variable with mean value λ = 1.2. The desired
probability is now obtained as follows:

P{X > 0} = 1 − P{X = 0} = 1 − [e−1.2(1.2)0 / 0!]

= 1 − e−1.2 = 0.6988