Probability and Statistics Seventh Week Text Book: E.

Kreyszig, Advanced Engineering Mathematics, John Wiley & Sons, Inc., 9th edition, 2006. Reference Books: M.R. Spiegel, J. Schiller and R.A. Srinivasan, Schaums Easy Outlines of Probability and Statistics, McGrawHill, 2001. R.E. Walpole, R.H. Myers, S.L. Myers and K. Ye, Probability & Statistics for Engineers & Scientists, Pearson Education, Inc. 8th edition, 2007. Outlines of the lecture Binomial, Poisson and Hypergeometric Distributions

Binomial Distribution

Binomial Distribution The binomial distribution occurs in games of chance, quality inspection, opinion polls, medicine, and so on. The conditions of its occurrence are as follows. We are interested in the number of times an event A occurs in n independent trials. In each trial the event A has the same probability P (A) = p. Then in a trial, A will not occur with probability q = 1 p. In n trials the random variable that interests us is

X = Number of times the event A occurs in n trials. X can assume the values 0, 1, , n, and we want to determine the corresponding probabilities. Now X = x means that A occurs in n trials and in n x trials it does not occur. This may look as follows. A A A( x terms,) B B B( n-x terms,) (1)

where B = Ac . We now use the assumption that the trials are independent, that is, they do not inuence each other. Hence (1) has the probability p p p( x terms,) q q q( n-x terms,) = px q nx (2)

We now use Theorem If n given things can be divided into c classes of alike things diering from class to class, then the number of permutations of these things taken all at a time is n! where n = n1 + n2 + + nc n1 !n2 ! nc ! where nj is the number of things in the jth class.

taking the number of permutations of n things (the n outcomes of the n trials) consisting of two classes, class one containing the x As and class two containing the n x Bs. This number is n! = n Cx x!(n x)! Accordingly, (2) multiplied by this binomial coecient n Cx gives the probability P (X = x) of X = x, that is, of obtaining A precisely x times in n trials. Hence X has the probability function f (x) = n Cx px q nx , x = 0, 1, 2, , n (3) and f (x) = 0 otherwise. The distribution of X with probability function (3) is called the binomial distribution or Bernoulli distribution. An experiment is called a binomial experiment if it possesses the following four properties: (1) The outcomes of each trial may be classied into one of the two categories, conventionally called success (A) and the failure (B). It is to be noted that the outcome of interest is called a success and the other, a failure. (2) The probability of success, denoted by p, remains constant for all trials. (3) The successive trials are all independent. (4) the experiment is repeated a xed number of times, say n. The mean of the binomial distribution is = np and the variance is 2 = npq (5) For the symmetric case of equal chance of success and failure (p = q = 1/2) this gives the mean n/2, the variance n/4, and the probability function
n

(4)

f (x) =

Cx , 2n

x = 0, 1, 2, , n

(6)

Example A fair coin is tossed 5 times. Find the probability of obtaining various number of heads. Solution Let us regard the tossing of a coin as an experiment. Then we observe that (i) each toss of a coin has two possible outcomes, heads and tails; (ii) the probability of a head is p = 1/2 and remains the same for successive tosses (iii) the successive tosses are independent; and (iv) the coin is tossed 5 times.

Therefore the random variable X which denotes the number of heads (successes) has a binomial probability distribution with p = q = 1/2 and n = 5. The possible values of X are 0, 1, 2, 3, 4 and 5. Hence P (X = 0) = P (X = 1) = P (X = 2) = P (X = 3) = P (X = 4) = P (X = 5) = C0 25 5 C1 25 5 C2 25 5 C3 25 5 C4 25 5 C5 25
5

= = = = = =

1 32 5 32 10 32 10 32 5 32 1 32

Example An event has the probability p = 3/8. Find the complete binomial distribution for n = 5 trials. Solution Here p = 3/8, q = 1 p = 5/8 and n = 5 P (X P (X P (X P (X P (X P (X = 0) = 5 C0 (3/8)0 (5/8)50 = 1) = 5 C1 (3/8)1 (5/8)51 = 2) = 5 C2 (3/8)2 (5/8)52 = 3) = 5 C3 (3/8)3 (5/8)53 = 4) = 5 C4 (3/8)4 (5/8)54 = 5) = 5 C5 (3/8)5 (5/8)55 = 0.0954 = 0.2861 = 0.3433 = 0.2060 = 0.0618 = 0.0074

Example Compute the probability of obtaining at least two Six in rolling a fair die 4 times. Solution p = p(A) = 1/6, q = 5/6, n = 4. The event At least two Six occurs if we obtain 2 or 3 or 4 Six. Hence the answer is P = f (2) + f (3) + f (4) 1 5 1 1 5 = 4 C2 ( )2 ( )2 + 4 C3 ( )3 ( ) + 4 C4 ( )4 = 0.132 6 6 6 6 6 Sampling with Replacement This mean, that we draw things from a given set one by one, and after each trial we replace the thing drawn (put it back to the given set and mix) before we draw the next thing. This guarantees independence of trials and leads to the binomial distribution. Indeed, if a box contains N things, for example, screws, M of which are defective, the probability of drawing a defective screw in a trial is p = M/N . Hence the probability of drawing a non-defective screw is q = 1 p = 1 M/N , and (3) gives the probability of drawing x defective in n trials in the form f (x) = n Cx ( M x M ) (1 )nx , N N x = 0, 1, , n (7)

Poisson Distribution

Poisson Distribution The discrete distribution with innitely many possible values and probability function x f (x) = e (8) x! It can be proved that this distribution is obtained as a limiting case of the binomial distribution, if we let p 0 and n so that the mean = np approaches a nite value. The Poisson distribution has the mean and the variance 2 = Example If the probability of producing a defective screw is p = 0.01 what is the probability that a lot of 100 screw will contain more than 2 defective? Solution The complementary event is Ac : Not more than 2 defective. Since p is very small, we can approximate this by the much more convenient Poisson distribution with mean = np = 100 0.01 = 1 1 P (Ac ) = e1 (1 + 1 + ) = 0.9197 2 Thus P (A) = 0.0803. Example If on the average 2 cars enter a certain parking lot per minute, what is the probability that during any given minute 4 or more car, will enter the lot? Solution To understand that the Poisson distribution is a model of the situation, we imagine the minute to be divided into very many short time intervals, let p be the (constant) probability that a car will enter the lot during any such short interval, and assume independence of the events that happen during those intervals. Then we are dealing with a binomial distribution with very large n and very small p, which we can approximate by the Poisson distribution with = np = 2 because 2 cars enter on the average. Solution The complementary event of the event 4 cars or more during a given minute is 3 cars or fewer enter the lot and has the probability P (Ac ) = f (0) + f (1) + f (2) + f (3) 21 22 23 20 = e2 + e2 + e2 + e2 = 0.857 0! 1! 2! 3! Hence P (A) = 0.143. Example If X has a poisson random variable with parameter = 2, nd the probabilities for x = 0, 1, 2, 3 or more. Solution Here the Poisson distribution is 2x 2 P (X = x) = e , x = 0, 1, 2, x! Hence 20 P (X = 0) = e2 = 0.13534 0! 21 P (X = 1) = e2 = 0.27067 1! 22 2 P (X = 2) = e = 0.27067 2!

Solution P (X 3) = 1 P (X < 3) = 1 (P (X = 0) + P (X = 1) + P (X = 2)) = 0.32333 Example Two hundred passengers have made reservations for an airplane ight. If the probability that a passenger who has a reservation will not show up is 0.01, what is the probability that exactly three will not show up? Solution Let us regard a no show as success. Then this is essentially a binomial experiment with n = 200 and p = 0.01. Since p is very small and n is considerably large, we shall apply the Poisson distribution, using = np = 200 0.01 = 2. Therefore, if X represent the number of successes, we have P (X = 3) = 23 e2 = 0.1804 3!

Hypergeometric Distribution Sampling without replacement means that we return no screw to the box. Then we no longer have independence of trials and instead of (7) the probability of drawing x defective in n trials is
M

f (x) =

Cx N M Cnx , NC n

x = 0, 1, 2, , n.

The distribution with this probability function is called the hypergeometric distribution. The hypergeometric distribution has the mean =n and the variance 2 = nM (N M )(N n) N 2 (N 1) M N

The hypergeometric probability distribution is appropriate when (1) A random sample of size n is drawn without replacement from a nite population of N units. (2) M of the units are of one kind (classied as success) and the remaining N M of another kind (classied as failure). Example An urn contains four red balls and six black balls. A sample of four balls is selected from the urn without replacement. Let X be the number red balls contained in the sample, then nd the probability distribution for X. Solution Here X is a hypergeometric random variable because (1) the results of each draw may be classied as either red (success) or black (failure), (2) the probability of success changes on each draw, (3) the successive draws are dependent as the selection is made without replacement, (4) the drawing is repeated a xed number of times (n=4).

Now, N = 4 + 6 = 10, M = 4, n = 4 and the possible values of X are 0, 1, 2, 3, 4. Therefore, the probabilities of these possible outcomes are
4

P (X = 0) = P (X = 1) = P (X = 2) = P (X = 3) = P (X = 4) =

C0 6 C4 10 C 4 4 6 C1 C3 10 C 4 4 6 C2 C2 10 C 4 4 C3 6 C1 10 C 4 4 C4 6 C0 10 C 4

= = = = =

15 210 80 210 90 210 24 210 15 210

Example We want to draw random samples of two gaskets from a box containing 10 gaskets, three of which are defective. Find the probability function of the random variable X = Number of defective in the sample. Solution For sampling without replacement we have to use hypergeometric distribution, nding
3

f (x) =

Cx 7 C2x 10 C 2 21 3 , f (2) = 45 45

f (0) = f (1) = Assignment

(i) Prove that Binomial distribution becomes Poisson distribution for p 0 and n (ii) Prove that variance of Hypergeometric distribution is 2 = nM (N M )(N n) N 2 (N 1)

(the assignment must be submitted after EID)