Chapter 5: Discrete Distributions 1 Chapter 5 Discrete Distributions LEARNING OBJECTIVES The overall learning objective of Chapter 5 is to help you understand a category of probability distributions that produces only discrete outcomes, thereby enabling you to: 1. Distinguish between discrete random variables and continuous random variables. 2. Know how to determine the mean and variance of a discrete distribution. 3. Identify the type of statistical experiments that can be described by the binomial distribution and know how to work such problems. 4. Decide when to use the Poisson distribution in analyzing statistical experiments and know how to work such problems. 5. Decide when binomial distribution problems can be approximated by the Poisson distribution and know how to work such problems. 6. Decide when to use the hypergeometric distribution and know how to work such problems . CHAPTER TEACHING STRATEGY Chapters 5 and 6 introduce the student to several statistical distributions. It is important to differentiate between the discrete distributions of chapter 5 and the continuous distributions of chapter 6. The approach taken in presenting the binomial distribution is to build on techniques presented in chapter 4. It can be helpful to take the time to apply the law of multiplication for independent events to a problem and demonstrate to students that sequence is important. From there, the student will more easily understand that by using combinations, one can more quickly determine the number of sequences and weigh the probability of obtaining a single sequence by that number. In a sense, we are developing the binomial formula through an inductive process. Chapter 5: Discrete Distributions 2 Thus, the binomial formula becomes more of a summary device than a statistical "trick". The binomial tables presented in this text are noncumulative. This makes it easier for the student to recognize that the table is but a listing of a series of binomial formula computations. In addition, it lends itself more readily to the graphing of a binomial distribution. It is important to differentiate applications of the Poisson distribution from binomial distribution problems. It is often difficult for students to determine which type of distribution to apply to a problem. The Poisson distribution applies to rare occurrences over some interval. The parameters involved in the binomial distribution (n and p) are different from the parameter (Lambda) of a Poisson distribution. It is sometimes difficult for students to know how to handle Poisson problems where the interval for the problem is different than the interval for which Lambda was developed. If they can view Lambda as a long run average which can be appropriately adjusted for various intervals, then students can be more successful with these types of problems. Solving for the mean and standard deviation of binomial distributions prepares the students for chapter 6 where the normal distribution is sometimes used to solve binomial distribution problems. In addition, graphing binomial and Poisson distributions affords the student the opportunity to visualize the meaning and impact of a particular set of parameters for a distribution. It can be emphasized that the hypergeometric distribution is an exact distribution. However, it is cumbersome to determine probabilities using the hypergeometric formula particularly when computing cumulative probabilities. The hypergeometric distribution can be presented as a fall-back position to be used when the binomial distribution should not be applied because of the non independence of trials and size of sample. CHAPTER OUTLINE 5.1 Discrete Versus Continuous Distributions 5.2 Describing a Discrete Distribution Mean, Variance, and Standard Deviation of Discrete Distributions Mean or Expected Value Variance and Standard Deviation of a Discrete Distribution 5.3 Binomial Distribution Solving a Binomial Problem Using the Binomial Table Using the Computer to Produce a Binomial Distribution Mean and Standard Deviation of the Binomial Distribution Graphing Binomial Distributions Chapter 5: Discrete Distributions 3 5.4 Poisson Distribution Working Poisson Problems by Formula Using the Poisson Tables Mean and Standard Deviation of a Poisson Distribution Graphing Poisson Distributions Using the Computer to Generate Poisson Distributions Approximating Binomial Problems by the Poisson Distribution 5.5 Hypergeometric Distribution Using the Computer to Solve for Hypergeometric Distribution Probabilities KEY TERMS Binomial Distribution Continuous Distributions Continuous Random Variables Discrete Distributions Discrete Random Variables Hypergeometric Distribution Lambda () Mean, or Expected Value Poisson Distribution Random Variable SOLUTIONS TO PROBLEMS IN CHAPTER 5 5.1 x P(x) x·P(x) 1 2 3 4 5 .238 .290 .177 .158 .137 .238 .580 .531 .632 .685 µ = [x·P(x)] = 2.666 = (x-µ)2 2.775556 0.443556 0.111556 1.779556 5.447556 (x-µ)2·P(x) 0.6605823 0.1286312 0.0197454 0.2811700 0.7463152 2 = (x-µ)2·P(x) = 1.836444 1.836444 = 1.355155 Chapter 5: Discrete Distributions 5.2 x P(x) x·P(x) 0 1 2 3 4 5 6 7 .103 .118 .246 .229 .138 .094 .071 .001 .000 .118 .492 .687 .552 .470 .426 .007 µ = [x·P(x)] = 2.752 = 5.3 x (x-µ)2 (x-µ)2·P(x) 7.573504 3.069504 0.565504 0.061504 1.557504 5.053504 10.549500 18.045500 0.780071 0.362201 0.139114 0.014084 0.214936 0.475029 0.749015 0.018046 2 = (x-µ)2·P(x) = 2.752496 2.752496 = 1.6591 P(x) 0 1 2 3 4 x·P(x) .461 .285 .129 .087 .038 .000 .285 .258 .261 .152 (x-µ)2 (x-µ)2·P(x) 0.913936 0.001936 1.089936 4.177936 9.265936 = 0.421324 0.000552 0.140602 0.363480 0.352106 2 = (x-µ)2·P(x) = 1.278064 E(x)=µ= [x·P(x)]= 0.956 5.4 4 1.278064 = 1.1305 x P(x) x·P(x) 0 1 2 3 4 5 6 .262 .393 .246 .082 .015 .002 .000 .000 .393 .492 .246 .060 .010 .000 µ = [x·P(x)] = 1.201 = .96260 = .98112 (x-µ)2 1.4424 0.0404 0.6384 3.2364 7.8344 14.4324 23.0304 (x-µ)2·P(x) 0.37791 0.01588 0.15705 0.26538 0.11752 0.02886 0.00000 2 = (x-µ)2·P(x) = 0.96260 Chapter 5: Discrete Distributions 5 5.5 a) n=4 P(x=3) = b) n=7 p = .10 3 1 4C3(.10) (.90) p = .80 q = .90 = 4(.001)(.90) = .0036 q = .20 P(x=4) = 7C4(.80)4(.20)3 = 35(.4096)(.008) = .1147 c) n = 10 p = .60 q = .40 P(x > 7) = P(x=7) + P(x=8) + P(x=9) + P(x=10) = 7 3 8 2 9 1 10 0 10C7(.60) (.40) + 10C8(.60) (.40) + 10C9(.60) (.40) +10C10(.60) (.40) 120(.0280)(.064) + 45(.0168)(.16) + 10(.0101)(.40) + 1(.0060)(1) = .2150 + .1209 + .0403 + .0060 = .3822 d) n = 12 p = .45 q = .55 P(5 < x < 7) = P(x=5) + P(x=6) + P(x=7) = 5 7 6 6 7 5 12C5(.45) (.55) + 12C6(.45) (.55) + 12C7(.45) (.55) = 792(.0185)(.0152) + 924(.0083)(.0277) + 792(.0037)(.0503) = .2225 + .2124 + .1489 = .5838 5.6 By Table A.2: a) n = 20 p = .50 P(x=12) = .120 b) n = 20 p = .30 P(x > 8) = P(x=9) + P(x=10) + P(x=11) + ...+ P(x=20) = = Chapter 5: Discrete Distributions 6 .065 + .031 + .012 + .004 + .001 + .000 = .113 c) n = 20 p = .70 P(x < 12) = P(x=11) + P(x=10) + P(x=9) + ... + (Px=0) = .065 + .031 + .012 + .004 + .001 + .000 = .113 d) n = 20 p = .90 P(x < 16) = P(x=16) + P(x=15) + P(x=14) + ...+ P(x=0) = .090 + .032 + .009 + .002 + .000 = .133 e) n = 15 p = .40 P(4 < x < 9) = P(x=4) + P(x=5) + P(x=6) + P(x=7) + P(x=8) + P(x=9) = .127 + .186 + .207 + .177 + .118 + .061 = .876 f) n = 10 p = .60 P(x > 7) = P(x=7) + P(x=8) + P(x=9) + P(x=10) = .215 + .122 + .040 + .006 = .382 5.7 a) n = 20 p = .70 q = .30 µ = np = 20(.70) = 14 = b) n = 70 n p q 20(.70)(.30) 4.2 = 2.05 p = .35 µ = np = 70(.35) = 24.5 q = .65 Chapter 5: Discrete Distributions = c) 7 n p q 70(.35)(.65) 15.925 = 3.99 n = 100 p = .50 q = .50 µ = np = 100(.50) = 50 = n p q 100(.50)(.50) 25 = 5 5.8 a) b) n=6 n = 20 p = .70 p = .50 x 0 1 2 3 4 5 6 Prob .001 .010 .060 .185 .324 .303 .118 x 0 1 2 3 4 5 6 7 Prob .000 .000 .000 .001 .005 .015 .037 .074 Chapter 5: Discrete Distributions c) n=8 p = .80 8 8 9 10 11 12 13 14 15 16 17 18 19 20 .120 .160 .176 .160 .120 .074 .037 .015 .005 .001 .000 .000 .000 x 0 1 2 3 4 5 6 7 8 Prob .000 .000 .001 .009 .046 .147 .294 .336 .168 Chapter 5: Discrete Distributions 9 5.9 a) n = 20 20C14 b) x = 14 (.78)14(.22)6 = 38,760(.030855)(.00011338) = .1356 n = 20 20C20 c) p = .78 p = .75 x = 20 (.75)20(.25)0 = (1)(.0031712)(1) = .0032 n = 20 p = .70 x < 12 Use table A.2: P(x=0) + P(x=1) + . . . + P(x=11)= .000 + .000 + .000 + .000 + .000 + .000 + .000 + .001 + .004 + .012 + .031 + .065 = .113 5.10 n = 16 p = .40 P(x > 9): from Table A.2: x Prob 9 10 11 12 13 .084 .039 .014 .004 .001 .142 Chapter 5: Discrete Distributions 10 P(3 < x < 6): x Prob 3 4 5 6 .047 .101 .162 .198 .508 n = 13 p = .88 P(x = 10) = 13C10(.88)10(.12)3 = 286(.278500976)(.001728) = .1376 P(x = 13) = 13C13(.88)13(.12)0 = (1)(.1897906171)(1) = .1898 Expected Value = µ = n p = 13(.88) = 11.44 5.11 n = 25 P = .60 a) x > 15 P(x > 15) = P(x = 15) + P(x = 16) + · · · + P(x = 25) Using Table A.2 x Prob 15 16 17 18 19 20 21 22 .161 .151 .120 .080 .044 .020 .007 .002 n = 25, p = .80 .585 b) x > 20 from a): P(x > 20) = P(x = 21) + P(x = 22) + P(x = 23) + P(x = 24) + P(x = 25) = .007 + .002 + .000 + .000 + .000 = .009 Chapter 5: Discrete Distributions 11 c) P(x < 10) from Table A.2, x Prob. 9 8 7 <6 .009 .003 .001 .000 .013 5.12 The highest probability values are for x = 15, 16, 14, 17, 13, 18, and 12. The expected value is 25(.60) = 15. The standard deviation is 2.45. 15 + 2(2.45) = 15 + 4.90 gives a range that goes from 10.10 to 19.90. From table A.2, the sum of the probabilities of the values in this range (11 through 19) is .936 or 93.6% of the values which compares quite favorably with the 95% suggested by the empirical rule. 5.13 n = 15 p = .20 a) P(x = 5) = 5 10 15C5(.20) (.80) = 3003(.00032)(.1073742) = .1032 b) P(x > 9): Using Table A.2 P(x = 10) + P(x = 11) + . . . + P(x = 15) = Chapter 5: Discrete Distributions .000 + .000 + . . . + .000 = .000 c) P(x = 0) = 0 15 15C0(.20) (.80) = (1)(1)(.035184) = .0352 d) P(4 < x < 7): Using Table A.2 P(x = 4) + P(x = 5) + P(x = 6) + P(x = 7) = .188 + .103 + .043 + .014 = .348 e) 5.14 n = 18 a) b) p =.30 µ = 18(.30) = 5.4 p = .34 µ = 18(.34) = 6.12 P(x > 8) n = 18 from Table A.2 x 8 9 10 11 12 Prob .081 .039 .015 .005 .001 .141 p = .30 12 Chapter 5: Discrete Distributions c) n = 18 13 p = .34 P(2 < x < 4) = P(x = 2) + P(x = 3) + P(x = 4) = 2 16 18C2(.34) (.66) + 3 15 18C3(.34) (.66) + 4 14 18C4(.34) (.66) = .0229 + .0630 + .1217 = .2076 d) n = 18 p = .30 0 18 18C0(.30) (.70) n = 18 = .00163 p = .34 0 18 18C0(.34) (.66) x=0 x=0 = .00056 The probability that none are in the $500,000 to $1,000,000 is higher because there is a smaller percentage in that category which is closer to zero. 5.15 a) Prob(x=5 = 2.3)= (2.35)(e-2.3) = (64.36343)(.1002588) = .0538 5! (120) b) Prob(x=2 = 3.9) = (3.92)(e-3.9) = (15.21)(.02024) = .1539 2! (2) c) Prob(x < 3 = 4.1) = Prob(x=3) + Prob(x=2) + Prob(x=1) + Prob(x=0) = (4.13)(e-4.1) = (68.921)(.016574) = .1904 3! 6 + (4.12)(e-4.1) = (16.81)(.016573) = .1393 2! 2 + (4.11)(e-4.1) = (4.1)(.016573) = .0679 1! 1 Chapter 5: Discrete Distributions 14 + (4.10)(e-4.1) = (1)(.016573) = .0166 0! 1 .1904 + .1393 + .0679 + .0166 = .4142 d) Prob(x=0 = 2.7) = (2.70)(e-2.7) = (1)(.0672) = .0672 0! 1 e) Prob(x=1 = 5.4)= (5.41)(e-5.4) = (5.4)(.0045) = .0244 1! 1 f) Prob(4 < x < 8 = 4.4) = Prob(x=5 = 4.4) + Prob(x=6 = 4.4) + Prob(x=7 = 4.4)= (4.45)(e-4.4) + (4.46)(e-4.4) + (4.47)(e-4.4) = 5! 6! 7! (1649.162)(.012277) + (7256.314)(.012277) + (31927.781)(.012277) 120 720 5040 = .1687 + .1237 + .0778 = .3702 5.16 a) Prob(x=6 = 3.8) = .0936 b) Prob(x>7 = 2.9): x 8 9 10 11 12 Prob .0068 .0022 .0006 .0002 .0000 x > 7 .0098 Chapter 5: Discrete Distributions 15 c) Prob(3 < x < 9 = 4.2)= x 3 4 5 6 7 8 9 3<x<9 Prob .1852 .1944 .1633 .1143 .0686 .0360 .0168 .7786 d) Prob(x=0 = 1.9) = .1496 e) Prob(x < 6 = 2.9)= x 0 1 2 3 4 5 6 x< 6 Prob .0050 .1596 .2314 .2237 .1622 .0940 .0455 .9214 f) Prob(5 < x < 8 = 5.7) = x 6 7 8 5<x < 8 5.17 a) = 6.3 Prob .1594 .1298 .0925 .3817 mean = 6.3 x 0 1 2 3 4 5 6 Standard deviation = Prob .0018 .0116 .0364 .0765 .1205 .1519 .1595 6.3 = 2.51 Chapter 5: Discrete Distributions 7 8 9 10 11 12 13 14 15 16 17 18 19 b) = 1.3 mean = 1.3 x 0 1 2 3 4 5 6 7 8 9 16 .1435 .1130 .0791 .0498 .0285 .0150 .0073 .0033 .0014 .0005 .0002 .0001 .0000 standard deviation = Prob .2725 .3542 .2303 .0998 .0324 .0084 .0018 .0003 .0001 .0000 1.3 = 1.14 Chapter 5: Discrete Distributions c) = 8.9 mean = 8.9 x 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 17 standard deviation = Prob .0001 .0012 .0054 .0160 .0357 .0635 .0941 .1197 .1332 .1317 .1172 .0948 .0703 .0481 .0306 .0182 .0101 .0053 .0026 .0012 .0005 .0002 .0001 8.9 = 2.98 Chapter 5: Discrete Distributions d) = 0.6 mean = 0.6 x 0 1 2 3 4 5 6 5.18 = 2.84 minutes a) Prob(x=6 = 2.8) from Table A.3 .0407 18 standard deviation = Prob .5488 .3293 .0988 .0198 .0030 .0004 .0000 0.6 = .775 Chapter 5: Discrete Distributions 19 b) Prob(x=0 = 2.8) = from Table A.3 .0608 c) Unable to meet demand if x > 44 minutes: x 5 6 7 8 9 10 11 x>4 .1523 Prob. .0872 .0407 .0163 .0057 .0018 .0005 .0001 .1523 probability of being unable to meet the demand. Probability of meeting the demand = 1 - (.1523) = .8477 15.23% of the time a second window will need to be opened. d) = 2.8 arrivals4 minutes Prob(x=3) arrivals2 minutes = ?? Lambda must be changed to the same interval (½ the size) New lambda=1.4 arrivals2 minutes Prob(x=3) =1.4) = from Table A.3 = .1128 Prob(x > 5 8 minutes) = ?? Lambda must be changed to the same interval(twice the size): New lambda= 5.6 arrivals8 minutes Prob(x > 5 = 5.6): Chapter 5: Discrete Distributions From Table A.3: x 5 6 7 8 9 10 11 12 13 14 15 16 17 x> 5 20 Prob. .1697 .1584 .1267 .0887 .0552 .0309 .0157 .0073 .0032 .0013 .0005 .0002 .0001 .6579 5.19 = x/n = 126/36 = 3.5 Using Table A.3 a) P(x = 0) = .0302 b) P(x > 6) = P(x = 6) + P(x = 7) + . . . = .0771 + .0385 + .0169 + .0066 + .0023 + .0007 + .0002 + .0001 = .1424 c) P(x < 4 10 minutes) = 7.0 10 minutes P(x < 4) = P(x = 0) + P(x = 1) + P(x = 2) + P(x = 3) = .0009 + .0064 + .0223 + .0521 = .0817 d) P(3 < x < 6 10 minutes) = 7.0 10 minutes P(3 < x < 6) = P(x = 3) + P(x = 4) + P(x = 5) + P(x = 6) = .0521 + .0912 + .1277 + .1490 = .42 Chapter 5: Discrete Distributions 21 e) P(x = 8 15 minutes) = 10.5 15 minutes P(x = 8 15 minutes) = 5.20 X e X! (10.58 )(e10.5 ) 8! = .1009 = 5.6 days3 weeks a) Prob(x=0 = 5.6): from Table A.3 = .0037 b) Prob(x=6 = 5.6): from Table A.3 = .1584 c) Prob(x > 15 = 5.6): x 15 16 17 x > 15 Prob. .0005 .0002 .0001 .0008 Because this probability is so low, if it actually occurred, the researcher would actually have to question the Lambda value as too low for this period. 5.21 = 0.6 trips 1 year a) Prob(x=0 = 0.6): from Table A.3 = .5488 b) Prob(x=1 = 0.6): from Table A.3 = .3293 c) Prob(x > 2 = 0.6): Chapter 5: Discrete Distributions from Table A.3 22 x 2 3 4 5 6 Prob. .0988 .0198 .0030 .0004 .0000 x > 2 .1220 d) Prob(x < 3 3 year period): The interval length has been increased (3 times) New Lambda = = 1.8 trips3 years Prob(x < 3 = 1.8): from Table A.3 x 0 1 2 3 x < 3 e) Prob(x=4 6 years): The interval has been increased (6 times) New Lambda = = 3.6 trips6 years Prob(x=4 = 3.6): from Table A.3 = .1912 5.22 = 1.2 collisions4 months a) Prob(x=0 = 1.2): from Table A.3 = .3012 b) Prob(x=2 2 months): Prob. .1653 .2975 .2678 .1607 .8913 Chapter 5: Discrete Distributions 23 The interval has been decreased (by ½) New Lambda = = 0.6 collisions2 months Prob(x=2 = 0.6): from Table A.3 = .0988 c) Prob (x < 1 collision6 months): The interval length has been increased (by 1.5) New Lambda = = 1.8 collisions6 months Prob(x < 1 = 1.8): from Table A.3 x 0 1 Prob. .1653 .2975 x< 1 .4628 The result is likely to happen almost half the time (46.26%). Ship channel and weather conditions are about normal for this period. Safety awareness is about normal for this period. There is no compelling reason to reject the lambda value of 0.6 collisions per 4 months based on an outcome of 0 or 1 collisions per 6 months. 5.23 = 1.2 penscarton a) Prob(x=0 = 1.2): from Table A.3 = .3012 b) Prob(x > 8 = 1.2): from Table A.3 = .0000 c) Prob(x > 3 = 1.2): from Table A.3 x 4 5 Prob. .0260 .0062 Chapter 5: Discrete Distributions 5.24 n = 100,000 24 6 7 8 .0012 .0002 .0000 x>3 .0336 p = .00004 Prob(x > 7 n = 100,000 p = .00004): = µ = np = 100,000(.00004) = 4.0 Since n > 20 and np < 7, the Poisson approximation to this binomial problem is close enough. Prob(x > 7 = 4): Using Table A.3 x 7 8 9 10 11 12 13 14 Prob. .0595 .0298 .0132 .0053 .0019 .0006 .0002 .0001 x > 7 .1106 x 11 12 13 14 Prob. .0019 .0006 .0002 .0001 x > 10 .0028 Prob(x>10 = 4): Using Table A.3 Since getting more than 10 is a rare occurrence, this particular geographic region appears to have a higher average rate than other regions. An investigation of particular characteristics of this region might be warranted. Chapter 5: Discrete Distributions 5.25 p = .009 25 n = 200 Use the Poisson Distribution: = np = 200(.009) = 1.8 P(x > 6) from Table A.3 = P(x = 6) + P(x = 7) + P(x = 8) + P(x = 9) + . . . = .0078 + .0020 + .0005 + .0001 = .0104 P(x > 10) = .0000 P(x = 0) = .1653 P(x < 5) = P(x = 0) + P(x = 1) + P(x = 2) + P( x = 3) + P(x = 4) = .1653 + .2975 + .2678 + .1607 + .0723 = .9636 5.26 n = 300, p = .01, = n(p) = 300(.01) = 3 a) Prob(x = 5): Using = 3 and Table A.3 = .1008 b) Prob (x < 4) = Prob.(x = 0) + Prob.(x = 1) + Prob.(x = 2) + Prob.(x = 3) = .0498 + .1494 + .2240 + .2240 = .6472 c) The expected number = µ = = 3 5.27 a) Prob(x = 3 N = 11, A = 8, n = 4) 8 C3 3 C1 (56)(3) = .5091 330 11 C 4 b) Prob(x < 2)N = 15, A = 5, n = 6) Prob(x = 1) + Prob (x = 0) = 5 C1 10 C5 + 15 C 6 5 C0 10 C6 (5)( 252) (1)( 210) = 5005 5005 15 C 6 Chapter 5: Discrete Distributions 26 .2517 + .0420 = .2937 c) Prob(x=0 N = 9, A = 2, n = 3) 2 C0 7 C3 (1)(35) = .4167 C 84 9 3 d) Prob(x > 4 N = 20, A = 5, n = 7) = Prob(x = 5) + Prob(x = 6) + Prob(x = 7) = 5 C5 15 C 2 + 20 C 7 5 C 6 15 C1 + 20 C 7 5 C7 15 C0 = 20 C 7 (1)(105) + 5C6 (impossible) + 5C7(impossible) = .0014 77520 5.28 N = 19 n = 6 a) P(x = 1 private) 11 A = 11 C1 8 C5 (11)(56) = .0227 27,132 19 C 6 b) P(x = 4 private) 11 C 4 8 C 2 (330)( 28) = .3406 C 27 , 132 19 6 c) P(x = 6 private) 11 C6 8 C0 (462)(1) = .0170 27,132 19 C 6 d) P(x = 0 private) 11 C0 8 C6 (1)( 28) = .0010 27,132 19 C 6 Chapter 5: Discrete Distributions 5.29 N = 17 A=8 8 a) P(x = 0) = b) P(x = 4) = 8 n=4 C 0 9 C 4 (1)(126) = = .0529 2380 17 C 4 C 4 9 C 0 (70)(1) = = .0294 2380 17 C 4 c) P(x = 2 non computer) = 5.30 N = 20 27 9 C 2 8 C 2 (36)( 28) = = .4235 2380 17 C 4 A = 16 white a) Prob(x = 4 white) = b) Prob(x = 4 red) = 4 16 N - A = 4 red C 4 4 C1 (1820)( 4) = 15504 20 C 5 n=5 = .4696 C 4 16 C1 (1)(16) = = .0010 15504 20 C 5 C5 16 C0 = .0000 because 4C5 is impossible to determine 20 C5 The participant cannot draw 5 red beads if there are only 4 to draw from. c) Prob(x = 5 red) = 5.31 N = 10 4 n=4 a) A = 3 x = 2 3 C 2 7 C 2 (3)( 21) = .30 210 10 C 4 b) A = 5 x = 0 5 C0 5 C 4 (1)(5) = .0238 210 10 C 4 c) A = 5 x = 3 5 C3 5 C1 (10)(5) = .2381 210 10 C 4 Chapter 5: Discrete Distributions 5.32 N = 16 A = 4 defective a) Prob(x = 0) = 4 b) Prob(x = 3) = 4 28 n=3 C0 12 C3 (1)( 220) = .3929 560 16 C 3 C3 12 C0 (4)(1) = .0071 560 16 C 3 c) Prob(x > 2) = Prob(x=2) + Prob(x=3) = 4 C 2 12 C1 + .0071 (from part b.) = 16 C 3 (6)(12) + .0071 = .1286 + .0071 = .1357 560 d) Prob(x < 1) = Prob(x=1) + Prob(x=0) = 4 5.33 C1 12 C 2 (4)(66) + .3929 (from part a.) = + .3929 = .4714 + .3929 = .8643 560 16 C 3 N = 18 A = 11 Hispanic n=5 Prob(x < 1) = Prob(1) + Prob(0) = 11 C1 7 C 4 + 18 C 5 11 C 0 7 C5 (11)(35) (1)( 21) = = .0449 + .0025 = .0474 8568 8568 18 C 5 It is fairly unlikely that these results occur by chance. A researcher might investigate further the causes of this result. Were officers selected based on leadership, years of service, dedication, prejudice, or what? 5.34 a) Prob(x=4 n = 11 and p = .23) 4 7 11C4(.23) (.77) = 330(.0028)(.1605) = .1482 b) Prob(x > 1n = 6 and p = .50) = 1 - Prob(x < 1) = 1 - Prob(x = 0) = 1-6C0(.50)0(.50)6 = 1-(1)(1)(.0156) = .9844 Chapter 5: Discrete Distributions 29 c) Prob(x > 7 n = 9 and p = .85) = Prob(x = 8) + Prob(x = 9) = 8 1 9C8(.85) (.15) + 9C9(.85)9(.15)0 = (9)(.2725)(.15) + (1)(.2316)(1) = .3679 + .2316 = .5995 d) Prob(x < 3 n = 14 and p = .70) = Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x = 0) = 3 11 14C3(.70) (.30) + 14C2(.70)2(.30)12 + 1 13 14C1(.70) (.30) + 14C0(.70)0(.30)14 = (364)(.3430)(.00000177) + (91)(.49)(.000000047)= (14)(.70)(.00000016) + (1)(1)(.000000047) = .0002 + .0000 + .0000 + .0000 = .0002 5.35 a) Prob(x = 14 n = 20 and p = .60) = .124 b) Prob(x < 5 n = 10 and p =.30) = Prob(x = 4) + Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x=0) = x 0 1 2 3 4 x<5 Prob. .028 .121 .233 .267 .200 .849 c) Prob(x > 12 n = 15 and p = .60) = Prob(x = 12) + Prob(x = 13) + Prob(x = 14) + Prob(x = 15) x 12 13 Prob. .063 .022 Chapter 5: Discrete Distributions 14 15 x > 12 30 .005 .000 .090 d) Prob(x > 20 n = 25 and p = .40) = Prob(x = 21) + Prob(x = 22) + Prob(x = 23) + Prob(x = 24) + Prob(x=25) = x 21 22 23 24 25 x > 20 Prob. .000 .000 .000 .000 .000 .000 5.36 a) Prob(x=4 = 1.25) (1.254)(e-1.25) = (2.4414)(.2865) = .0291 4! 24 b) Prob(x < 1 = 6.37) = Prob(x = 1) + Prob(x = 0) = (6.37)1(e-6.37) + (6.37)0(e-6.37) = (6.37)(.0017) + (1)(.0017) = 1! 0! 1 1 .0109 + .0017 = .0126 c) Prob(x > 5 = 2.4) = Prob(x = 6) + Prob(x = 7) + ... = (2.4)6(e-2.4) + (2.4)7(e-2.4) + (2.4)8(e-2.4) + (2.4)9(e-2.4) + (2.4)10(e-2.4) + ... 6! 7! 8! 9! 10! .0241 + .0083 + .0025 + .0007 + .0002 = .0358 for values x > 11 the probabilities are each .0000 when rounded off to 4 decimal places. Chapter 5: Discrete Distributions 5.37 31 a) Prob(x = 3 = 1.8) = .1607 b) Prob(x < 5 = 3.3) = Prob(x = 4) + Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x = 0) = x 0 1 2 3 4 x<5 Prob. .0369 .1217 .2008 .2209 .1823 .7626 c) Prob(x > 3 = 2.1) = x 3 4 5 6 7 8 9 10 11 x> 5 Prob. .1890 .0992 .0417 .0146 .0044 .0011 .0003 .0001 .0000 .3504 d) Prob(2 < x < 5 = 4.2) = Prob(x=3) + Prob(x=4) + Prob(x=5) = 5.38 x 3 4 5 Prob. .1852 .1944 .1633 2<x <5 .5429 a) Prob(x = 3 N = 6, n = 4, A = 5) = 5 C3 1 C1 (10)(1) = .6667 C 15 6 4 Chapter 5: Discrete Distributions 32 b) Prob(x < 1 N = 10, n = 3, A = 5) = Prob(x = 1) + Prob(x = 0) = 5 C1 5 C 2 + 10 C 3 5 C 0 5 C 3 (5)(10) (1)(10) = 120 120 10 C 3 = .4167 + .0833 = .5000 c) Prob(x > 2 N = 13, n = 5, A = 3) = Prob(x=2) + Prob(x=3) Note: only 3 x's in population 3 C 2 10 C3 + 13 C 5 3 C3 10 C 2 (3)(120) (1)( 45) = 1287 1287 13 C 5 = .2797 + .0350 = .3147 5.39 n = 25 p = .20 retired from Table A.2: P(x = 7) = .111 P(x > 10): P(x = 10) + P(x = 11) + . . . + P(x = 25) = .012 + .004 + .001 = .017 Expected Value = µ = np = 25(.20) = 5 n = 20 p = .40 mutual funds P(x = 8) = .180 P(x < 6) = P(x = 0) + P(x = 1) + . . . + P(x = 5) = .000 + .000 + .003 +.012 + .035 + .075 = .125 P(x = 0) = .000 P(x > 12) = P(x = 12) + P(x = 13) + . . . + P(x = 20) = .035 + .015 + .005 + .001 = .056 x=8 Expected Number = µ = n p = 20(.40) = 8 Chapter 5: Discrete Distributions 5.40 33 = 3.2 cars2 hours a) Prob(x=3) cars per 1 hour) = ?? The interval has been decreased by ½. The new = 1.6 cars1 hour. Prob(x = 3 = 1.6) = (from Table A.3) .1378 b) Prob(x = 0 cars per ½ hour) = ?? The interval has been decreased by ¼ the original amount. The new = 0.8 cars½ hour. Prob(x = 0 = 0.8) = (from Table A.3) .4493 c) Prob(x > 5 = 1.6) = (from Table A.3) x 5 6 7 8 Prob. .0176 .0047 .0011 .0002 .0236 Either a rare event occurred or perhaps the long-run average, , has changed (increased). 5.41 N = 32 A = 10 a) P(x = 3) = 10 b) P(x = 6) = 10 c) P(x = 0) = 10 n = 12 C3 22 C9 (120)( 497,420) = = .2644 225,792,840 32 C12 C6 22 C6 (210)(74,613) = = .0694 225,792,840 32 C12 C0 22 C12 (1)(646,646) = = .0029 225,792,840 32 C12 Chapter 5: Discrete Distributions 34 d) A = 22 P(7 < x < 9) = = 22 C7 10 C5 + 32 C12 22 C8 10 C 4 + 32 C12 22 C9 10 C3 32 C12 (170,544)( 252) (319,770)( 210) (497,420)(120) 225,792,840 225,792,840 225,792,840 = .1903 + .2974 + .2644 = .7521 5.42 = 1.4 defects 1 lot If x > 3, buyer rejects If x < 3, buyer accepts Prob(x < 3 = 1.4) = (from Table A.3) 5.43 x 0 1 2 3 Prob. .2466 .3452 .2417 .1128 x<3 .9463 a) n = 20 and p = .25 The expected number = µ = np = (20)(.25) = 5.00 b) Prob(x < 1 n = 20 and p = .25) = Prob(x = 1) + Prob(x = 0) = 20C1(.25)1(.75)19 + 20C0(.25)0(.75)20 = (20)(.25)(.00423) + (1)(1)(.0032) = .0212 +. 0032 = .0244 Since the probability is so low, the population of your state may have a lower percentage of chronic heart conditions than those of other states. 5.44 a) Prob(x > 7 n = 10 and p = .70) = (from Table A.2): x 8 9 10 Prob. .233 .121 .028 x>7 .382 Chapter 5: Discrete Distributions 35 Expected number = µ = np = 10(.70) = 7 b) Expected number = µ = np = 15 (1/3) = 5 Prob(x=0 n = 15 and p = 1/3) = 0 15 15C0() () = .0023 c) Prob(x = 7 n = 7 and p = .53) = 7C7(.53)7(.47)0 = .0117 Probably the 53% figure is too low for this population since the probability of this occurrence is so low (.0117). 5.45 n = 12 a.) Prob(x = 0 long hours): p = .20 0 12 12C0(.20) (.80) = .0687 b.) Prob(x > 6) long hours): p = .20 Using Table A.2: .016 + .003 + .001 = .020 c) Prob(x = 5 good financing): p = .25, 5 7 12C5(.25) (.75) = .1032 d.) p = .19 (good plan), expected number = µ = n(p) = 12(.19) = 2.28 5.46 n = 100,000 p = .000014 Worked as a Poisson: = np = 100,000(.000014) = 1.4 a) P(x = 5): from Table A.3 = .0111 b) P(x = 0): Chapter 5: Discrete Distributions 36 from Table A.3 = .2466 c) P(x > 6): x 7 8 5.47 (from Table A.3) Prob .0005 .0001 .0006 Prob(x < 3) n = 8 and p = .60) = (from Table A.2) x 0 1 2 3 x< 3 Prob. .001 .008 .041 .124 .174 17.4% of the time in a sample of eight, three or fewer customers are walk-ins by chance. Other reasons for such a low number of walk-ins might be that she is retaining more old customers than before or perhaps a new competitor is attracting walk-ins away from her. 5.48 n = 25 p = .20 a) Prob(x = 8 n = 25 and p = .20) = (from Table A.2) .062 b) Prob(x > 10) n=25 and p = .20) = (from Table A.2) x 11 12 13 x > 10 Prob. .004 .001 .000 .005 c) Since such a result would only occur 0.5% of the time by chance, it is likely that the analyst's list was not representative of the entire state of Idaho or the 20% figure for the Idaho census is not correct. 5.49 = 0.6 flats2000 miles Prob(x = 0 = 0.6) = (from Table A.3) .5488 Chapter 5: Discrete Distributions 37 Prob(x > 3 = 0.6) = (from Table A.3) x 3 4 5 Prob. .0198 .0030 .0004 x>3 One trip is independent of the other. Let F = flat tire and NF = no flat tire P(NF1 NF2) = P(NF1) P(NF2) P(NF) = .5488 P(NF1 NF2) = (.5488)(.5488) = .3012 5.50 N = 25 n=8 a) P(x = 1 in NY) 4 A=4 C1 21 C7 (4)(116,280) = = .4300 1,081,575 25 C8 b) P(x = 4 in top 10) 10 A = 10 C 4 15 C 4 (210(1365) = .2650 1,081,575 25 C8 c) P(x = 0 in California) 4 C0 21 C8 (1)( 203,490) = .1881 1,081,575 25 C8 d) P(x = 3 with M) 3 5.51 A=4 A=3 C3 22 C5 (1)( 26,334) = .0243 1,081,575 25 C8 N = 24 n=6 A=8 .0232 Chapter 5: Discrete Distributions a) P(x = 6) = b) P(x = 0) = 8 38 C6 16 C0 (28)(1) = .0002 C 134 , 596 24 6 8 C0 16 C6 (1)(8008) = .0595 134,596 24 C 6 d) A = 16 East Side P(x = 3) = 5.52 16 C3 8 C3 (560)(56) = .2330 134,596 24 C 6 n = 25 p = .20 Expected Value = µ = np = 25(.20) = 5 µ = 25(.20) = 5 = n p q 25(.20)(.80) = 2 P(x > 12) = (from Table A.2) x 13 Prob .0000 The values for x > 12 are so far away from the expected value that they are very unlikely to occur. P(x = 14) = 25C14(.20)14(.80)11 = .000063 which is very unlikely. If this value (x = 14) actually occurred, one would doubt the validity of the p = .20 figure or one would have experienced a very rare event. Chapter 5: Discrete Distributions 5.53 39 = 2.4 calls1 minute a) Prob(x = 0 = 2.4) = (from Table A.3) .0907 b) Can handle x < 5 calls Cannot handle x > 5 calls Prob(x > 5 = 2.4) = (from Table A.3) x 6 7 8 9 10 11 x>5 c) Prob(x = 3 calls 2 minutes) The interval has been increased 2 times. New Lambda = = 4.8 calls2 minutes. from Table A.3: .1517 d) Prob(x < 1 calls15 seconds): The interval has been decreased by ¼. New Lambda = = 0.6 calls15 seconds. Prob(x < 1 = 0.6) = (from Table A.3) Prob(x = 1) = .3293 Prob(x = 0) = .5488 Prob(x < 1) = .8781 Prob. .0241 .0083 .0025 .0007 .0002 .0000 .0358 Chapter 5: Discrete Distributions 5.54 n = 160 40 p = .01 Working this problem as a Poisson problem: a) Expected number = µ = n(p) = 160(.01) = 1.6 b) P(x > 8): Using Table A.3: x 8 9 Prob. .0002 .0000 .0002 x 2 3 4 5 6 P(2 < x < 6) Prob. .2584 .1378 .0551 .0176 .0047 .4736 c) P(2 < x < 6): Using Table A.3: 5.55 p = .005 n = 1,000 = np = (1,000)(.005) = 5 a) P(x < 4) = P(x = 0) + P(x = 1) + P(x = 2) + P(x = 3) = .0067 + .0337 + .0842 + .1404 = .265 b) P(x > 10) = P(x = 11) + P(x = 12) + . . . = .0082 + .0034 + .0013 + .0005 + .0002 = .0136 c) P(x = 0) = .0067 Chapter 5: Discrete Distributions 5.56 n=8 P = .36 0 8 8C0(.36) (.64) 41 x = 0 Women = (1)(1)(.0281475) = .0281 It is unlikely that a company would randomly hire 8 physicians from the U.S. pool and none of them would be female. If this actually happened, the figures might be used as evidence in a lawsuit. 5.57 N = 32 a) n = 5 10 A = 10 C3 22 C 2 (120)( 231) = .1377 201,376 32 C5 b) n = 8 6 x=3 x<2 C0 26 C8 + 32 C8 A=6 6 C1 26 C7 + 32 C8 6 C 2 26 C 6 = 32 C8 (1)(1,562,275) (6)(657,800) (15)(38,760) = .1485 + .3752 + .0553 = .5790 10,518,300 10,518,300 10,518,300 c) n = 5 x=2 p = 3/26 = .1154 2 3 5C2(.1154) (.8846) 5.58 N = 14 = (10)(.013317)(.692215) = .0922 n=4 a) P(x = 4 N = 14, n = 4, A = 10 Northside) 10 C 4 4 C0 (210((1) = .2098 1001 14 C 4 b) P(x = 4 N = 14, n = 4, A = 4 West) 4 C 4 10 C0 (1)(1) = .0010 1001 14 C 4 c) P(x = 2 N = 14, n = 4, A = 4 West) Chapter 5: Discrete Distributions 4 5.59 42 C 2 10 C 2 (6)( 45) = .2697 C 1001 14 4 a) = 3.841,000 3.840 e3.84 P(x = 0) = = .0215 0! b) = 7.682,000 P(x = 6) = 7.686 e7.68 (205,195.258)(.000461975) = .1317 6! 720 c) = 1.61,000 and = 4.83,000 from Table A.3: P(x < 7) = P(x = 0) + P(x = 1) + . . . + P(x = 6) = .0082 + .0395 + .0948 + .1517 + .1820 + .1747 + .1398 = .7907 5.60 This is a binomial distribution with n = 15 and p = .36. = np = 15(.36) = 5.4 = 15(.36)(.64) = 1.86 The most likely values are near the mean, 5.4. Note from the printout that the most probable values are at x = 5 and x = 6 which are near the mean. 5.61 This printout contains the probabilities for various values of x from zero to eleven from a Poisson distribution with = 2.78. Note that the highest probabilities are at x = 2 and x = 3 which are near the mean. The probability is slightly higher at x = 2 than at x = 3 even though x = 3 is nearer to the mean because of the “piling up” effect of x = 0. 5.62 This is a binomial distribution with n = 22 and p = .64. The mean is np = 22(.64) = 14.08 and the standard deviation is: = n p q 22(.64)(.36) = 2.25 Chapter 5: Discrete Distributions 43 The x value with the highest peak on the graph is at x = 14 followed by x = 15 and x = 13 which are nearest to the mean. 5.63 This is the graph of a Poisson Distribution with = 1.784. Note the high probabilities at x = 1 and x = 2 which are nearest to the mean. Note also that the probabilities for values of x > 8 are near to zero because they are so far away from the mean or expected value.