Professional Documents
Culture Documents
Discrete Distributions: Learning Objectives
Discrete Distributions: Learning Objectives
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:
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
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 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.
CHAPTER OUTLINE
KEY TERMS
σ = 1.836444 = 1.355155
Chapter 5: Discrete Distributions 4
σ = 2.752496 = 1.6591
σ = 1.278064 = 1.1305
σ = .96260 = .98112
Chapter 5: Discrete Distributions 5
5.5
a) n=4 p = .10 q = .90
3 1
P(x=3) = 4C3(.10) (.90) = 4(.001)(.90) = .0036
c) n = 10 p = .60 q = .40
d) n = 12 p = .45 q = .55
a) n = 20 p = .50
P(x=12) = .120
b) n = 20 p = .30
c) n = 20 p = .70
d) n = 20 p = .90
e) n = 15 p = .40
f) n = 10 p = .60
5.7
a) n = 20 p = .70 q = .30
µ = n⋅p = 20(.70) = 14
b) n = 70 p = .35 q = .65
µ = n⋅p = 100(.50) = 50
σ = n ⋅ p ⋅ q = 100(.50)(.50) = 25 = 5
5.8
a) n = 6 p = .70 x Prob
0 .001
1 .010
2 .060
3 .185
4 .324
5 .303
6 .118
b) n = 20 p = .50 x Prob
0 .000
1 .000
2 .000
3 .001
4 .005
5 .015
6 .037
7 .074
Chapter 5: Discrete Distributions 8
8 .120
9 .160
10 .176
11 .160
12 .120
13 .074
14 .037
15 .015
16 .005
17 .001
18 .000
19 .000
20 .000
5.9
a) n = 20 p = .78 x = 14
b) n = 20 p = .75 x = 20
c) n = 20 p = .70 x < 12
5.10 n = 16 p = .40
x Prob
9 .084
10 .039
11 .014
12 .004
13 .001
.142
Chapter 5: Discrete Distributions 10
x Prob
3 .047
4 .101
5 .162
6 .198
.508
n = 13 p = .88
5.11 n = 25 P = .60
a) x > 15
x Prob
15 .161
16 .151
17 .120
18 .080
19 .044
20 .020
21 .007
22 .002
.585
b) x > 20
from a): P(x > 20) = P(x = 21) + P(x = 22) + P(x = 23) +
9 .009
8 .003
7 .001
<6 .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
5 10
a) P(x = 5) = 15C5(.20) (.80) =
3003(.00032)(.1073742) = .1032
(1)(1)(.035184) = .0352
e)
5.14 n = 18
x Prob
8 .081
9 .039
10 .015
11 .005
12 .001
.141
Chapter 5: Discrete Distributions 13
c) n = 18 p = .34
d) n = 18 p = .30 x=0
0 18
18C0(.30) (.70) = .00163
n = 18 p = .34 x=0
0 18
18C0(.34) (.66) = .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.
b) Prob(x=2λ = 3.9) =
d) Prob(x=0λ = 2.7) =
e) Prob(x=1 λ = 5.4)=
b) Prob(x>7 λ = 2.9):
x Prob
8 .0068
9 .0022
10 .0006
11 .0002
12 .0000
x > 7 .0098
Chapter 5: Discrete Distributions 15
x Prob
3 .1852
4 .1944
5 .1633
6 .1143
7 .0686
8 .0360
9 .0168
3<x<9 .7786
x Prob
0 .0050
1 .1596
2 .2314
3 .2237
4 .1622
5 .0940
6 .0455
x< 6 .9214
x Prob
6 .1594
7 .1298
8 .0925
5<x < 8 .3817
x Prob
0 .0018
1 .0116
2 .0364
3 .0765
4 .1205
5 .1519
6 .1595
Chapter 5: Discrete Distributions 16
7 .1435
8 .1130
9 .0791
10 .0498
11 .0285
12 .0150
13 .0073
14 .0033
15 .0014
16 .0005
17 .0002
18 .0001
19 .0000
x Prob
0 .2725
1 .3542
2 .2303
3 .0998
4 .0324
5 .0084
6 .0018
7 .0003
8 .0001
9 .0000
Chapter 5: Discrete Distributions 17
x Prob
0 .0001
1 .0012
2 .0054
3 .0160
4 .0357
5 .0635
6 .0941
7 .1197
8 .1332
9 .1317
10 .1172
11 .0948
12 .0703
13 .0481
14 .0306
15 .0182
16 .0101
17 .0053
18 .0026
19 .0012
20 .0005
21 .0002
22 .0001
Chapter 5: Discrete Distributions 18
x Prob
0 .5488
1 .3293
2 .0988
3 .0198
4 .0030
5 .0004
6 .0000
a) Prob(x=6 λ = 2.8)
b) Prob(x=0 λ = 2.8) =
x Prob.
5 .0872
6 .0407
7 .0163
8 .0057
9 .0018
10 .0005
11 .0001
x>4 .1523
a) P(x = 0) = .0302
λ = 7.0 10 minutes
e) P(x = 8 15 minutes)
λ = 10.5 15 minutes
a) Prob(x=0 λ = 5.6):
b) Prob(x=6 λ = 5.6):
x Prob.
15 .0005
16 .0002
17 .0001
x > 15 .0008
a) Prob(x=0 λ = 0.6):
b) Prob(x=1 λ = 0.6):
x > 2 .1220
x < 3 .8913
e) Prob(x=4 6 years):
Prob(x=4 λ = 3.6):
a) Prob(x=0 λ = 1.2):
b) Prob(x=2 2 months):
Chapter 5: Discrete Distributions 23
Prob(x=2 λ = 0.6):
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.
a) Prob(x=0 λ = 1.2):
6 .0012
7 .0002
8 .0000
x>3 .0336
Since n > 20 and n⋅p < 7, the Poisson approximation to this binomial problem is
close enough.
x > 7 .1106
Prob(x>10 λ = 4):
x > 10 .0028
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 25
P(x = 0) = .1653
a) Prob(x = 5):
C3 ⋅3 C1 (56)(3)
8
= = .5091
11 C 4 330
Prob(x = 1) + Prob (x = 0) =
c) Prob(x=0 N = 9, A = 2, n = 3)
C 0 ⋅7 C 3 (1)(35)
2
= = .4167
9 C3 84
(1)(105)
+ 5C6 (impossible) + 5C7(impossible) = .0014
77520
5.28 N = 19 n = 6
a) P(x = 1 private) A = 11
C1 ⋅8 C 5 (11)(56)
11
= = .0227
19 C 6 27,132
b) P(x = 4 private)
C 4 ⋅8 C 2 (330)(28)
11
= = .3406
19 C 6 27,132
c) P(x = 6 private)
C 6 ⋅8 C 0 (462)(1)
11
= = .0170
19 C 6 27,132
d) P(x = 0 private)
C 0 ⋅8 C 6 (1)(28)
11
= = .0010
19 C 6 27,132
Chapter 5: Discrete Distributions 27
8 C 0 ⋅9 C 4 (1)(126)
a) P(x = 0) = = = .0529
17 C 4 2380
8 C 4 ⋅9 C 0 (70)(1)
b) P(x = 4) = = = .0294
17 C 4 2380
9 C 2 ⋅8 C 2 (36)(28)
c) P(x = 2 non computer) = = = .4235
17 C 4 2380
16 C 4 ⋅ 4 C1 (1820)(4)
a) Prob(x = 4 white) = = = .4696
20 C 5 15504
4 C 4 ⋅16 C1 (1)(16)
b) Prob(x = 4 red) = = = .0010
20 C 5 15504
C 5 ⋅16 C 0
4
c) Prob(x = 5 red) = = .0000 because 4C5 is impossible to determine
20 C 5
The participant cannot draw 5 red beads if there are only 4 to draw from.
5.31 N = 10 n=4
a) A = 3 x = 2
C 2 ⋅7 C 2 (3)(21)
3
= = .30
10 C 4 210
b) A = 5 x = 0
C 0 ⋅5 C 4 (1)(5)
5
= = .0238
10 C 4 210
c) A = 5 x = 3
C 3 ⋅5 C1 (10)(5)
5
= = .2381
10 C 4 210
Chapter 5: Discrete Distributions 28
C 0 ⋅12 C3 (1)(220)
a) Prob(x = 0) = 4
= = .3929
16 C 3 560
C 3 ⋅12 C 0 (4)(1)
b) Prob(x = 3) = 4
= = .0071
16 C 3 560
4 C 2 ⋅12 C1
c) Prob(x > 2) = Prob(x=2) + Prob(x=3) = + .0071 (from part b.) =
16 C 3
4 C1 ⋅12 C 2 (4)(66)
+ .3929 (from part a.) = + .3929 = .4714 + .3929 = .8643
16 C 3 560
C1 ⋅7 C 4 C 0 ⋅7 C5 (11)(35) (1)(21)
11
+ 11
= + = .0449 + .0025 = .0474
18 C 5 18 C 5 8568 8568
(364)(.3430)(.00000177) + (91)(.49)(.000000047)=
(14)(.70)(.00000016) + (1)(1)(.000000047) =
x Prob.
0 .028
1 .121
2 .233
3 .267
4 .200
x<5 .849
x Prob.
12 .063
13 .022
Chapter 5: Discrete Distributions 30
14 .005
15 .000
x > 12 .090
x Prob.
21 .000
22 .000
23 .000
24 .000
25 .000
x > 20 .000
5.36
a) Prob(x=4 λ = 1.25)
for values x > 11 the probabilities are each .0000 when rounded off to 4
decimal places.
Chapter 5: Discrete Distributions 31
x Prob.
0 .0369
1 .1217
2 .2008
3 .2209
4 .1823
x<5 .7626
x Prob.
3 .1890
4 .0992
5 .0417
6 .0146
7 .0044
8 .0011
9 .0003
10 .0001
11 .0000
x> 5 .3504
x Prob.
3 .1852
4 .1944
5 .1633
5.38 a) Prob(x = 3 N = 6, n = 4, A = 5) =
C 3 ⋅1 C1 (10)(1)
5
= = .6667
6 C4 15
Chapter 5: Discrete Distributions 32
C1 ⋅5 C 2 C 0 ⋅5 C 3 (5)(10) (1)(10)
5
+ 5
= +
10 C 3 10 C 3 120 120
P(x = 8) = .180
P(x = 0) = .000
P(x > 12) = P(x = 12) + P(x = 13) + . . . + P(x = 20) =
x=8
x Prob.
5 .0176
6 .0047
7 .0011
8 .0002
.0236
Either a rare event occurred or perhaps the long-run average, λ, has changed
(increased).
5.41 N = 32 A = 10 n = 12
10 C 3 ⋅ 22 C 9 (120)(497,420)
a) P(x = 3) = = = .2644
32 C12 225,792,840
10 C 6 ⋅ 22 C 6 (210)(74,613)
b) P(x = 6) = = = .0694
32 C12 225,792,840
10 C 0 ⋅ 22 C12 (1)(646,646)
c) P(x = 0) = = = .0029
32 C12 225,792,840
Chapter 5: Discrete Distributions 34
d) A = 22
5.42 λ = 1.4 defects 1 lot If x > 3, buyer rejects If x < 3, buyer accepts
x Prob.
0 .2466
1 .3452
2 .2417
3 .1128
x<3 .9463
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.
x Prob.
8 .233
9 .121
10 .028
x>7 .382
Chapter 5: Discrete Distributions 35
0 15
15C0() () = .0023
Probably the 53% figure is too low for this population since the probability of
this occurrence is so low (.0117).
5.45 n = 12
p = .20
a) P(x = 5):
b) P(x = 0):
Chapter 5: Discrete Distributions 36
x Prob
7 .0005
8 .0001
.0006
x Prob.
0 .001
1 .008
2 .041
3 .124
x< 3 .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
x Prob.
11 .004
12 .001
13 .000
x > 10 .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.
x Prob.
3 .0198
4 .0030
5 .0004
x>3 .0232
P(NF) = .5488
P(NF1 _ NF2) = (.5488)(.5488) = .3012
5.50 N = 25 n=8
4 C1 ⋅ 21 C 7 (4)(116,280)
= = .4300
25 C 8 1,081,575
C 4 ⋅15 C 4 (210(1365)
10
= = .2650
25 C 8 1,081,575
C 0 ⋅ 21 C8 (1)(203,490)
4
= = .1881
25 C 8 1,081,575
C 3 ⋅ 22 C 5 (1)(26,334)
3
= = .0243
25 C 8 1,081,575
C 6 ⋅16 C 0 (28)(1)
a) P(x = 6) = 8
= = .0002
24 C 6 134,596
C 0 ⋅16 C 6 (1)(8008)
b) P(x = 0) = 8
= = .0595
24 C 6 134,596
d) A = 16 East Side
C 3 ⋅8 C 3 (560)(56)
P(x = 3) = 16
= = .2330
24 C 6 134,596
5.52 n = 25 p = .20
µ = 25(.20) = 5
σ = n ⋅ p ⋅ q = 25(.20)(.80) = 2
P(x > 12) = (from Table A.2)
x Prob
13 .0000
The values for x > 12 are so far away from the expected value that they are very
unlikely to occur.
x Prob.
6 .0241
7 .0083
8 .0025
9 .0007
10 .0002
11 .0000
x>5 .0358
Prob(x = 1) = .3293
Prob(x = 0) = .5488
Prob(x < 1) = .8781
Chapter 5: Discrete Distributions 40
λ = n⋅p = (1,000)(.005) = 5
c) P(x = 0) = .0067
Chapter 5: Discrete Distributions 41
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 x=3 A = 10
C 3 ⋅ 22 C 2 (120)(231)
10
= = .1377
32 C 5 201,376
b) n = 8 x < 2 A=6
6 C 0 ⋅ 26 C8 6 C1 ⋅ 26 C 7 6 C 2 ⋅ 26 C 6
+ + =
32 C 8 32 C 8 32 C 8
c) n = 5 x = 2 p = 3/26 = .1154
2 3
5C2(.1154) (.8846) = (10)(.013317)(.692215) = .0922
5.58 N = 14 n = 4
C 4 ⋅ 4 C 0 (210((1)
10
= = .2098
14 C 4 1001
C 4 ⋅10 C 0 (1)(1)
4
= = .0010
14 C 4 1001
C 2 ⋅10 C 2 (6)(45)
4
= = .2697
14 C 4 1001
5.59 a) λ = 3.841,000
3.840 ⋅ e −3.84
P(x = 0) = = .0215
0!
b) λ = 7.682,000
5.60
σ = 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 λ = 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.
The mean is n⋅p = 22(.64) = 14.08 and the standard deviation is:
σ = n ⋅ p ⋅ q = 22(.64)(.36) = 2.25
Chapter 5: Discrete Distributions 43
5.63 This is the graph of a Poisson Distribution with λ = 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.