Page 110 - Applied Statistics And Probability For Engineers
P. 110

PQ220 6234F.Ch 03  13/04/2002  03:19 PM  Page 88






               88     CHAPTER 3 DISCRETE RANDOM VARIABLES AND PROBABILITY DISTRIBUTIONS

                                 the probability that more than 45 of the sampled customers have purchased from the corpora-
                                 tion in the last three months?
                                    The sampling is without replacement. However, because the sample size of 50 is small
                                 relative to the number of customer accounts, 1000, the probability of selecting a customer who
                                 has purchased from the corporation in the last three months remains approximately constant
                                 as the customers are chosen.
                                    For example, let  A denote the event that the first customer selected has purchased
                                 from the corporation in the last three months, and let B denote the event that the second
                                 customer selected has purchased from the corporation in the last three months.  Then,
                                 P1A2   700 1000   0.7  and P1B ƒ A2   699 999   0.6997 . That is, the trials are approxi-
                                 mately independent.
                                    Let X denote the number of customers in the sample who have purchased from the cor-
                                 poration in the last three months. Then, X is a hypergeometric random variable with N
                                 1000, n   50, and K   700. Consequently,  p   K N   0.7 . The requested probability is
                                 P1X   452 . Because the sample size is small relative to the batch size, the distribution of X
                                 can be approximated as binomial with n   50 and p   0.7. Using the binomial approximation
                                 to the distribution of X results in

                                                             50  50
                                                                       x
                                                P1X   452    a   a  b 0.7 11   0.72 50 x    0.00017
                                                            x 46  x
                                 The probability from the hypergeometric distribution is 0.000166, but this requires computer
                                 software. The result agrees well with the binomial approximation.



               EXERCISES FOR SECTION 3-8
               3-86.  Suppose  X has a hypergeometric distribution with  (d) What is the mean number of unacceptable washers in the
               N   100, n   4, and K   20. Determine the following:  sample?
               (a) P1X   12  (b) P1X   62                      3-91.  A company employs 800 men under the age of 55.
               (c) P1X   42  (d) Determine the mean and variance of X.  Suppose that 30% carry a marker on the male chromosome
               3-87.  Suppose  X has a hypergeometric distribution with  that indicates an increased risk for high blood pressure.
               N   20, n   4, and K   4. Determine the following:  (a) If 10 men in the company are tested for the marker in this
               (a) P1X   12  (b) P1X   42                         chromosome, what is the probability that exactly 1 man
               (c) P1X   22  (d) Determine the mean and variance of X.  has the marker?
               3-88.  Suppose  X has a hypergeometric distribution with  (b) If 10 men in the company are tested for the marker in this
               N   10, n   3, and K   4. Sketch the probability mass func-  chromosome, what is the probability that more than 1 has
               tion of X.                                         the marker?
               3-89.  Determine the cumulative distribution function for X  3-92.  Printed circuit cards are placed in a functional test
               in Exercise 3-88.                               after being populated with semiconductor chips. A lot contains
               3-90.  A lot of 75 washers contains 5 in which the variability  140 cards, and 20 are selected without replacement for func-
                                                               tional testing.
               in thickness around the circumference of the washer is unac-
               ceptable.  A sample of 10 washers is selected at random,  (a) If 20 cards are defective, what is the probability that at
               without replacement.                               least 1 defective card is in the sample?
               (a) What is the probability that none of the unacceptable  (b) If 5 cards are defective, what is the probability that at least
                  washers is in the sample?                       1 defective card appears in the sample?
               (b) What is the probability that at least one unacceptable  3-93.  Magnetic tape is slit into half-inch widths that are
                  washer is in the sample?                     wound into cartridges. A slitter assembly contains 48 blades.
               (c) What is the probability that exactly one unacceptable  Five blades are selected at random and evaluated each day for
                  washer is in the sample?
   105   106   107   108   109   110   111   112   113   114   115