- BINOMIAL PROBABILITY DISTRIBUTION
- POISSON PROBABILITY DISTRIBUTON

Probability distribution deals with theoretical probability model based on the randomness of certain natural occurrences. The binomial and Poisson distribution are discrete distribution

**BINOMIAL DISTRIBUTION**

This arises from a repeated random experiment which has two possible outcomes.

The two possible outcomes of the random experiment are usually called success and failure.

Prob( success) = P, Prob(failure) = q

Since the two events are complementary, hence p+ q = 1 or p = 1-q, q = 1 – p

The probability of success or failure of an event is the same for each trials and does not influence the probability of success or failure of another trial of the same event.

:. Binomial distribution of n trails and r required outcome(s) is defined as :

pr(x = r) = ^{n}C_{r}P^{r}q^{n-r}

when^{n}C_{r} = n!

(n-r)! r!.

The binomial distribution is suitable when the number of trials is not too large.

**Example:**

1. Find the probability that when two fair coins are tossed 5 times a head and a tail appear three times.

Solution:

Two fair coins = (HT, TH, TT, HH) = 4

Prob (a head and a tail) = ^{2}/_{4} = ½

i.e p = ½ , q = ½ (p + q = 1)

n = 5, r = 3.

:. P(x = r) =^{n}Crp^{r}q^{n-r}

p ( x = 3) = ^{5}C_{3} ( ½ )^{ 3} ( ½ ) ^{5-3}

p (x = 3) = 10 x ^{1}/_{8} x ¼ = ^{10}/_{32} = ^{5}/_{16}

p (x = 3) = 0.3125.

**EVALUATION**

Find the probability that when a fair six-faced die is tossed six times, a prime number appears exactly four times.

**POISSON DISTRIBUTION**: The Poisson distribution is more suitable when the number of trials is very large and probability of successes is small. It is defined as:

Pr(x) = λ^{x} e^{– λ} , x = 0, 1, 2, 3,

x!

Where λ = np e = 2.718

P = probability of success, n = number of trials.

**EVALUATION**

The probability that a person gets a reaction from a new drug on the market is 0.001. If 200 people are treated with this drug. Find approximately, the probability that:

- exactly three persons will get a reaction
- more than two person will get a reaction

Properties of Binomial and Poisson Distribution.

**Binomial**

It assigns probability to non-occurrence of events i.eProb( x = 0)

Mean µ = np

Standard deviation, r = √npq

Variance ð^{2} = npq

**Poisson**

It assigns probability to non-occurrence of events i.eProb(x = 0)

Mean µ = λ = np

Standard deviation, ð = √λ = √np

Variance, ð^{2} = λ = np

Example:

1. In the probability of tossing a fair coin three times, a head shows up twice. Find the mean and standard deviation.

**Solution**

n = 3 Prob(a head) = ½ , i.e p = ½ , q = ½

I. Mean µ = np = 3 x ½ = 3/2

II. Standard deviation :r = √npq = 3 x ½ x ½ = ¾

Example

2. 0.2% of the cooks produced by a machine were found to be defective. If there are 1000 corks, find the mean and standard deviation?

**Solution**

P = 0.2% = 0.002.

N = 1000

I. mean µ = λ = np = 1000 x 0.002 = 2

II. r = √np = √2

**EVALUATION**

In an examination, 60% of the candidates pass. If 10 candidates were sampled. Find the mean, standard deviation and variance of the candidates.

**GENERAL EVALUATION**

- The probability that a person gets a reaction from a new drug in the market is 0.001. If 2000 people were treated with this drug, find the mean and standard deviation.
- 1. In an examination, 60% of the candidates passed. Use the binomial distribution to calculate the probabilities that a random sample of 10 candidates contain exactly 2 failures.

**READING ASIGNMENT**

Read probability distribution, further math. Project 3 from page 198-201.

**WEEKEND ASSIGNMENT**

1. What is the variance of a binomial distribution?

(a) np (b) √npq ( c ) npq (d) p^{2}

2. The mean (µ) of a poisson distribution is the same as

(a) Standard deviation (b) variance (c) mean (d) mean deviation

3. If number of trials is 100 and probability of success is 0.0001, what is the variance of this distribution?

(a) 0.00999 (b) 0.1 (c ) 0.01 (d) 0.001

4. If the birth of a male child and that of a female child are equiprobable. Find the probability that in a family of five children exactly 3 will be male. (a) 16/5 (b) 5/16 (c) 5/32 (d) 5/21

5. If an unbiased die is thrown repeatedly, what are the chances that the first, six to be thrown will be the third throw? (a) 25/216 (b) 1/6 (c) 25/36 (d)25/31

**THEORY**

1. 20% of the total production of transistors produced by a machine are below standard. If a random sample of 6 transistors produced by the machine is taken, what is the probability of getting

(i) exactly 2 (ii) exactly 1 (iii) at least 2 (iv) at most 2 standard transistors?

2. A fair die is thrown five times. Calculate correct to 3 decimal places, the probability of obtaining

(a) at most two sixes (b) exactly three sixes