Binomial Distribution

Section 7.2 Binomial Distribution

Consider a sequence of n independent Bernoulli trials with the likelihood of a success p on each individual trial stays constant from trial to trial with

0 < p < 1 .

If we let the variable

X

measure the number of successes obtained when doing a fixed number of trials n with

R = {0, 1, . . ., n},

then the resulting distribution of probabilities is called a Binomial Distribution.

🔗

Now, let’s determine the actual probability function for this distribution.

🔗

Theorem 7.2.1. Binomial Probability Function.

🔗

For R = {0, 1, ..., n},

f (x) = (\binom{n}{x}) p^{x} (1 - p)^{n - x}

🔗

Theorem 7.2.2. Verification of Binomial Distribution Formula.

🔗

\sum_{x \in R} f (x) = \sum_{x = 0}^{n} (\binom{n}{x}) p^{x} (1 - p)^{n - x} = 1.

🔗

Example 7.2.3. Flipping a coin a fixed number of times.

Let’s consider a simple example for flipping coins. Indeed, suppose you flip a coin exactly 20 times and need to determine the probability of getting exactly 10 heads.

This is binomial with n = 20, p = 1/2 and you are looking for f(10). With these values

f (10) = (\binom{20}{10}) \cdot {(\frac{1}{2})}^{10} \cdot {(\frac{1}{2})}^{20 - 10} = \frac{46189}{262144} \approx 0.176

Note that the mean for this distribution is also 10 so one might expect 10 heads in general

If you rather would prefer to determine the probability of getting 10 or fewer heads requires F(10) = f(0) + f(1) + ... + f(10). There is no "nice" formula for F but this calculation can be performed using a graphing calculator, such as the TI-84 with F(x) = binomcdf(n,p,x). In this case, F(10) = binomcdf(20,1/2,10) = 0.588.

🔗

Checkpoint 7.2.4. WebWork - Binomial.

A coin is tossed 11 times.

a) How many different outcomes are possible?

b) What is the probability of getting exactly 4 heads?

c) What is the probability of getting at least 2 heads?

d) What is the probability of getting at most 7 heads?

Answer 1.

2048

Answer 2.

0.1611328125

Answer 3.

0.994140625

Answer 4.

0.88671875

🔗

Example 7.2.5. Testing critical components.

Often one will test a critical system components for failure and toward that end collect a sample of 100 of these components from the manufacturer. Suppose the component is listed as having a p = 0.01 probability of breaking and you want to know the likelihood that at most 1 of the tested components actually fails when tested. You find it reasonable to presume that different components succeed or fail independently of each other. So, you can model this situation with a binomial distribution.

X

measures the number of components that fail when tested, the specific probability function is given by

f (x) = (\binom{100}{x}) {0.01}^{x} {0.99}^{100 - x} .

The probability that at most one component fails is then given by

F (1) = f (0) + f (1) = (\binom{100}{0}) {0.01}^{0} {0.99}^{100} + (\binom{100}{1}) {0.01}^{1} {0.99}^{9} 9 = {0.99}^{100} + 100 \cdot 0.01 \cdot {0.99}^{99} = {0.99}^{99} (0.99 + 100 \cdot 0.01 = {0.99}^{99} \cdot 1.99 \approx 0.73576 .

🔗

Utilize the interactive cells below to compute

f (x)

and

F (x)

for the Binomial distribution


    
        
xxxxxxxxxx
 
1
# Binomial calculator
2
@interact
3
def _(p=input_box(3/10,width=15),n=input_box(10,width=15)):
4
    R = range(n+1)
5
    f(x) = binomial(n,x)*p^x*(1-p)^(n-x)
6
    acc = 0
7
    print("Binomial Calculator")
8
    for k in R:
9
        prob = f(x=k)
10
        acc = acc+prob
11
        print('f(%s) = '%k,' %.8f'%prob,
12
                     ' and F(%s) = '%k,' %.8f'%acc)

    
    
    
    
        
            
                Language:
                
            
        
    
    




    
    
        
        Messages

🔗

Here is another calculator that you might like!


    
        
xxxxxxxxxx
 
1
# Calculator for Binomial 
2
print("Binomial Distribution Calculator")
3
pretty_print(html("n = fixed number of trials"))
4
pretty_print(html("p = probability of success on each trial"))
5
@interact
6
def _(n = input_box(default=10,width=10,label="n"),
7
      p = input_box(default=0.3,width=10,label="p"),
8
      x=input_box(default=4,width=10,label="x")):
9
    prob = binomial(n,x)*p^x*(1-p)^(n-x)
10
    pretty_print(html("$P(X=%s)$"%str(x)+"$=%s$"%str(prob)))
11
    binoms = [binomial(n,k)*p^k*(1-p)^(n-k) for k in range(x+1)]
12
    G = bar_chart([binomial(n,k)*p^k*(1-p)^(n-k) for k in range(n+1)]) # show all
13
    G += bar_chart(binoms,color='green')                                        # show till x
14
    show(G,figsize=(3,2))
15
    F = 0   #  accumulate probabilities from the bottom till x
16
    for k in range(x+1):
17
        F += binomial(n,k)*p^k*(1-p)^(n-k)
18
    end
19
    pretty_print(html("$P(X \\le %s)$"%str(x)+"$=%s$"%str(F)))

    
    
    
    
        
            
                Language:
                
            
        
    
    




    
    
        
        Messages

🔗

Theorem 7.2.6. Binomial Distribution Statistics.

🔗

μ = n p

σ^{2} = n p (1 - p)

γ_{1} = \frac{1 - 2 p}{\sqrt{n p (1 - p)}}

γ_{2} = \frac{1 - 6 p (1 - p)}{n p (1 - p)} + 3

🔗

Proof.

For the mean 1,

\begin{aligned} μ & = E [X] \\ = \sum_{x = 0}^{n} x (\binom{n}{x}) p^{x} (1 - p)^{n - x} \\ = \sum_{x = 1}^{n} x \frac{n (n - 1)!}{x (x - 1)! (n - x)!} p^{x} (1 - p)^{n - x} \\ = n p \sum_{x = 1}^{n} \frac{(n - 1)!}{(x - 1)! ((n - 1) - (x - 1))!} p^{x - 1} (1 - p)^{(n - 1) - (x - 1)} \end{aligned}

Using the change of variables

k = x - 1

and

m = n - 1

yields a binomial series

\begin{aligned} = n p \sum_{k = 0}^{m} \frac{m!}{k! (m - k)!} p^{k} (1 - p)^{m - k} \\ = n p (p + (1 - p))^{m} = n p \end{aligned}

For the variance 2,

\begin{matrix} σ^{2} = E [X (X - 1)] + μ - μ^{2} \\ = \sum_{x = 0}^{n} x (x - 1) (\binom{n}{x}) p^{x} (1 - p)^{n - x} + n p - n^{2} p^{2} \\ = \sum_{x = 2}^{n} x (x - 1) \frac{n (n - 1) (n - 2)!}{x (x - 1) (x - 2)! (n - x)!} p^{x} (1 - p)^{n - x} + n p - n^{2} p^{2} \\ = n (n - 1) p^{2} \sum_{x = 2}^{n} \frac{(n - 2)!}{(x - 2)! ((n - 2) - (x - 2))!} p^{x - 2} (1 - p)^{(n - 2) - (x - 2)} + n p - n^{2} p^{2} \end{matrix}

Using the change of variables

k = x - 2

and

m = n - 2

yields a binomial series

\begin{matrix} = n (n - 1) p^{2} \sum_{k = 0}^{m} \frac{m!}{k! (m - k)!} p^{k} (1 - p)^{m - k} + n p - n^{2} p^{2} \\ = n (n - 1) p^{2} + n p - n^{2} p^{2} = n p - n p^{2} = n p (1 - p) \end{matrix}

The skewness 3 and kurtosis 4 can be found similarly using formulas involving E[X(X-1)(X-2)] and E[X(X-1)(X-2)(X-3)]. The complete determination is performed using Sage below.

🔗

The following uses Sage to symbolically confirm the general formulas for the Binomial distribution.


    
        
xxxxxxxxxx
 
1
var('x,n,p')
2
assume(x,'integer')
3
f(x) = binomial(n,x)*p^x*(1-p)^(n-x)
4
mu = sum(x*f,x,0,n)
5
M2 = sum(x^2*f,x,0,n)
6
M3 = sum(x^3*f,x,0,n)
7
M4 = sum(x^4*f,x,0,n)
8
9
print('Mean = ',mu)
10
11
v = (M2-mu^2).factor()
12
print('Variance = ',v)
13
stand = sqrt(v)
14
15
sk = ((M3 - 3*M2*mu + 2*mu^3)).factor()/stand^3
16
print('Skewness = ',sk)
17
18
kurt = (M4 - 4*M3*mu + 6*M2*mu^2 -3*mu^4).factor()/stand^4
19
print('Kurtosis = ',(kurt-3).factor(),'+3')

    
    
    
    
        
            
                Language:
                
            
        
    
    




    
    
        
        Messages

🔗

Theorem 7.2.7. Binomial Limiting Distribution.

🔗

For the binomial distribution 7.2.1, as

m \to \infty

γ_{1} \to 0

🔗

and

γ_{2} \to 3.

🔗

Proof.

For skewness, take the limit of the skewness result above

lim_{n \to \infty} \frac{n (2 p - 1) (p - 1) p}{(- n (p 1) p)^{3 / 2}} = lim_{n \to \infty} C \frac{n}{n^{3 / 2}} = 0.

Similarly for kurtosis

lim_{n \to \infty} - \frac{6 p^{2} - 6 p + 1}{n (p - 1) p} + 3 = 0 + 3 = 3.

🔗

Checkpoint 7.2.8. WebWork - Binomial.

The Census Bureau reports that 82% of Americans over the age of 25 are high school graduates. A survey of randomly selected residents of certain county included 1260 who were over the age of 25, and 1086 of them were high school graduates.

(a)

Find the mean and standard deviation for the number of high school graduates in groups of 1260 Americans over the age of 25.

Mean =

Standard deviation =

(b)

Is that county result of 1086 unusually high, or low, or neither?

(Enter HIGH or LOW or NEITHER)

Answer 1.

1033.2

Answer 2.

13.6373017859106

Answer 3.

high


    
        
xxxxxxxxxx
 
1
n = 10
2
p = 0.3
3
X = 0:n    # the space R of the random variable 
4
mu = n*p      # the formula for mean of the Binomial Distributions
5
sdev = sqrt(n*p*(1-p))  # the formula for the standard deviation
6
if(n < 101){
7
dbinom( X, n, p )   # print out a bunch of actual probs if N reasonable
8
}
9
10
Pbinom = dbinom(X, n, p )  # create the probability function over X
11
12
Psample = rbinom(10^6, n, p)  # to create a histogram, sample a lot
13
Xtop=max(Psample)          # for scaling the x-axis. Shift by 1/2 below.
14
hist(Psample, prob=TRUE, br=(-1:Xtop)+0.5, col="skyblue2", xlab="X", 
15
  main="Binomial Probability Function vs Approximating 'Bell Curve'")
16
17
points(X, Pbinom, pch=19, col="darkgreen")  # to create actual (x,f(x))
18
19
Pnormal <- function(X){dnorm(X, mean=mu, sd=sdev)}   # overlap bell curve
20
curve(Pnormal, col="red", lwd=2, add=TRUE)

    
    
    
    
        
            
                Language:
                
            
        
    
    




    
    
        
        Messages

🔗

You can of course get specific values and graph the Binomial Distribution using R as well...


    
        
xxxxxxxxxx
 
1
n <- 10
2
p <- 0.3
3
4
paste('Probability Function')
5
dbinom(0:n, n, p)   # gives the probability function
6
paste('Distribution function')
7
pbinom(0:n, n, p)   # gives the distribution function
8
paste('A random sample')
9
rbinom(15, n, p)    # gives a random sample of 15 items from b(n,p)
10
11
x <- dbinom(0:n, size=n, prob=p)
12
barplot(x,names.arg=0:n, main=sprintf(paste('n=',n,' and p= ',p)))

    
    
    
    
        
            
                Language:
                
            
        
    
    




    
    
        
        Messages

Essentials of Mathematical Probability and Statistics

Search Results:

Section 7.2 Binomial Distribution

Theorem 7.2.1. Binomial Probability Function.

Proof.

Theorem 7.2.2. Verification of Binomial Distribution Formula.

Proof.

Example 7.2.3. Flipping a coin a fixed number of times.

Checkpoint 7.2.4. WebWork - Binomial.

Example 7.2.5. Testing critical components.

Theorem 7.2.6. Binomial Distribution Statistics.

Proof.

Theorem 7.2.7. Binomial Limiting Distribution.

Proof.

Checkpoint 7.2.8. WebWork - Binomial.