Professional Documents
Culture Documents
UNIVERSITY OF LEEDS
January 2016
ANSWERS
QUESTION ANSWER QUESTION ANSWER
1 A 31 E
2 A 32 A
3 B 33 A
4 C 34 D
5 E 35 C
6 C 36 D
7 E 37 C
8 B 38 A
9 B 39 E
10 B 40 C
11 D 41 E
12 A 42 C
13 D 43 A
14 B 44 D
15 D 45 C
16 D 46 A
17 C 47 B
18 D 48 C
19 D 49 D
20 B 50 C
21 D 51 C
22 B 52 B
23 A 53 A
24 E 54 C
25 D 55 C
26 A 56 A
27 A 57 B
28 B 58 A
29 C 59 C
30 C 60 B
2
1. We can use the following property of a joint probability distribution:
f x, y 1
x y
In our case f x, y c 1
1 2 1
y x x y2 , x 0, 2, 3, y 0, 1, 3, 5 . So we
21 7
have
3 5
1 1 2
c 1 21 y
x 0 y 0
2
x
7
xy 1
3 5
1 2 1 3 5
1 1 2
x 0 y 0
c 1
21
y x x y2 c
7 x 0
1 21 y
y 0
2
x
7
xy
3 5 1 3 5 3 5
1 3 5
c 1 y2 x x y 2
x 0 y 0 21 x 0 y 0 x 0 y 0 7 x 0 y 0
3 1 3 3
1 3
c 41 35 4 x x 35
x 0 21 x 0 x 0 7 x 0
3 1 3 3 3
c 4 5 4 x 5 x
x 0 3 x 0 x 0 x 0
c 34 35 45 55 c 12 5 20 25 62 c
1
3
1
62 c 1 c
62
1
f x, y
1 2 1
1 y x x y2 , x 0, 2, 3, y 0, 1, 3, 5 .
62 21 7
ANSWER: A
3
f x f x, y for x x1 , x 2 , ..., x n
y
In our case f x, y c 1
1 2 1
y x x y2 , x 0, 2, 3, y 0, 1, 3, 5 . So we
21 7
have
5
f x c 1
1 2 1
y x x y2
y 0 21 7
ANSWER: A
5
f x c 1
1 2 1
3. y x x y2
y 0 21 7
5
1 1 2 1
1 y x x y2
y 0 62 21 7
1 5 1 5 5
1 5
1 y 2 x x y 2
62 y 0 21 y 0 y 0 7 y 0
1 1
4 1 35 4 x x 35
1 1 5
4 4 x 5x
62 21 7 62 3
1 17 2 17 9
9x x
62 3 186 62
f x
17 9
x, x 0, 2, 3
186 62
ANSWER: B
f y f x, y for y y1 , y 2 , ..., y m
x
4
In our case f x, y c 1
1 2 1
y x x y2 , x 0, 2, 3, y 0, 1, 3, 5 . So we
21 7
have
3
f y c 1
1 2 1
y x x y2
x 0 21 7
ANSWER: C
3
f y c 1
1 2 1
5. y x x y2
x 0 21 7
3
1 1 2 1
1 y x x y2
x 0 62 21 7
1 3 1 3 3
1 3
1 y2 x y2 x
62 x 0 21 x 0 x 0 7 x 0
1 6 2 1 3 2 4 3 2
8 y 4 y y
62 7 31 7 31 217
f y
4 3 2
y , y 0, 1, 3, 5
31 217
ANSWER: E
f X 0, y
f y X 0
f X 0
In our case, f x, y y x x y 2 and f x
1 1 2 1 17 9
1 x . So we
62 21 7 186 62
have
5
2 1 1 2
1 21 y 0 7 0 y 62 1 21 y
1 1 2 1
f X 0, y
f y X 0
62
f X 0 17
9
0 1 17
186 62 62 3
3 1 2 3 31 3 1 1
1 y y2 y2
17 21 17 17 21 17 17 7
3 1 2
y
17 119
f y X 0
3 1 2
y , y 0, 1, 3, 5
17 119
ANSWER: C
5 5
4 3 2
7. E Y y f y y y
y 0 y 0 31 217
5
4 3 3 4 5 3 5 3
y y 217
y y
y 0 31 217 31 y 0 y 0
4
9 3 153 36 459 252 459 711 3.276
31 217 31 217 217 217 217
ANSWER: E
3 1 2
E Y X 0 y f y X 0 y
5 5
8. y
y 0 y 0 17 119
5
3 1 3 3 5 1 5 3
y y y y
y 0 17 119 17 y 0 119 y 0
3
9 1 153 27 153 189 153 342 2.874
17 119 17 119 119 119 119
ANSWER: B
6
9. In general, the probability density function (pdf ) of a continuous random variable X
following a (negative) exponential distribution is
f x e x , x 0, 0
EX
1 1
then
EX
In our case the mean (expected value) of patient waiting time is E X 1.4 ,
therefore:
1 1
0.714
E X 1.4
So the pdf of X is
ANSWER: B
Since e 2.718... 0 , the expression e 0.714 x is also positive for any value x .
Therefore
f x 0 x, x 0
i.e. f x is positive.
df x de g x
f ' x 0.714 0.714 e 0.714 x 0.714 2 e 0.714 x recall : g ' x e g x
dx dx
Therefore
f ' x 0 x, x 0
7
i.e. f x is decreasing.
d 2 f x
f ' ' x 0.714 2 0.714 e 0.714 x 0.714 3 e 0.714 x
dx 2
Therefore
f ' ' x 0 x, x 0
i.e. f x is convex.
0.714
x f x 0.714 e
1
0.714 0
e
0 x
ANSWER: B
SD X var X
1 1
2
In our case:
SD X
1
1.4
ANSWER: D
12. The mean number of patients treated per unit of time, i.e. the mean number of patients
treated per hour, is the mean occurrence rate:
8
1
EX
In our case:
1 1
0.714 mean number of patients treated per hour
E X 1.4
ANSWER: A
b
Pa X b f x dx
a
In our case, the probability that patient waiting time will be longer than 4 hours is
4
P X 4 1 P 0 X 4 1 e
x
dx
x 0
4 4
4
1 0.714
1
e 0.714 x 1 e 0.714 x
4
0 recall: e
cx
dx
1 cx
e
0.714 0 c
1 e 0.714 4 e 0.714 0 1 e 2.857 e 0
1 0.057 1 0.057
ANSWER: D
5 5 5
14. P 4 X 5 e dx
x
0.714 e
0.714 x
dx 0.714 e 0.714 x dx
x 4 x 4 x 4
9
5
0.714
1
e 0.714 x e 0.714 x 4
5
0.714 4
e 0.714 5 e 0.714 4 e 3.571 e 2.857
ANSWER: B
15. If X and Y are continuous random variables and f x, y is their joint probability
distribution, the general definition of marginal distribution of X is
d
f x f x, y dy for a xb
y c
0.15
f x x y 2 x 2 y 4 dy
125
y 0.15
9
ANSWER: D
0.15
f x x y 2 x 2 y 4 dy
125
16.
y 0.15
9
0.15 0.15
125
x 2 y 2 dy 125 x 2 y 2 dy
y 0.15
9 9 y 0.15
0.15
125
x 2 2 y
y2
9 2 0.15
x 2 0.15 20.15 ( 0.15) 2( 0.15)
2 2
125
9 2 2
10
x 2 0.15 0.30 0.15 0.30 125 x 2 0.60
2 2
125
9 2 2 9
75
x 2 25 x 2
9 3
f x x 2 ,
25
0.03 x 0.03
3
ANSWER: D
0.03
17. EX x f x dx
x 0.03
0.03
25 x 3 x2
0.03 0.03
x x 2 dx
25 25
2
3 3 2 0.03
x 2 x dx 2
x 0.03
3 3 x 0.03
25 0.033 2 ( 0.03) 3 2
0.03
( 0.03)
3 3 3
25 0.033 0.033
0.03
2
0.032
3 3 3
0.00135
0.00015
9
ANSWER: C
E X 2
0.03
18. x 2 f x dx
x 0.03
0.03
25 x 4 x3
x 2 x dx 3 4 2 3
0.0.3 0.03
x x 2 dx
25 25
2 3 2
x 0.03
3 3 x 0.03 0.03
11
25 0.034 0.033 ( 0.03) 4 ( 0.03) 3
2 2
3 4 3 4 3
0.0027
0.00030
9
ANSWER: D
19.
var X E X 2 E X
2
0.0002999775 0.00030
ANSWER: D
20. If X and Y are continuous random variables and f x, y is their joint probability
distribution, the general definition of marginal distribution of Y is
b
f y f x, y dx for c yd
xa
0.03
f y x y 2 x 2 y 4 dx
125
x 0.03
9
ANSWER: B
0.03
f y x y 2 x 2 y 4 dx
125
21.
x 0.03
9
12
0.03 0.03
125
x 2 y 2 dx 125 y 2 x 2 dx
x 0.03
9 9 x 0.03
0.03
y 2 x 2 x
2
125
9 2 0.03
y 2 0.03 20.03 ( 0.03) 2( 0.03)
2 2
125
9 2 2
15
y 2 5 y 2
9 3
f y y 2 ,
5
0.15 y 0.15
3
ANSWER: D
0.15
22. E Y y f y dy
y 0.15
0.15
5 y3 y2
0.15 0.15
y y 2 dy
5 5
2
y 2 y dy 3 2
y 0.15
3 3 y 0.15
3 2 0.15
5 0.153 2 ( 0.15) 3
0 .15
( 0.15) 2
3 3 3
5 0.153 0.153
0.152 0.152
3 3 3
0.03375
0.00375
9
13
ANSWER: B
E Y
0.15
23. 2
y 2 f y dy
y 0.15
0.15
5 y4 y3
0.15 0.15
y y 2 dy
5 5
2 3 2
y 2 y dy 4 2
y 0.15
3 3 y 0.15
3 3 0.15
0.0675
0.00750
9
ANSWER: A
var Y E Y 2 E Y
2
24.
0.0074859375 0.00749
ANSWER: E
25. If X and Y are continuous random variables and f x, y is their joint probability
distribution, the general definition of expected value of the product of X and Y is
b d
E XY xy f x, y dy dx
x a y c
14
In our case, f x, y x y 2 x 2 y 4 , 0.03 x 0.03 , 0.15 y 0.15 .
125
9
So we have
0.03 0.15
E XY x y 2 x 2 y 4 dy dx
125
x 0.03
y 0.15
xy
9
x y 2 x 2 y 2 xy 2 4 xy dy dx
0.03 0.15
125 2 2
x 0.03
y 0.15
9
ANSWER: D
26. Let us first check for the possible independence of the random variables X and Y .
To check for independence of X and Y , it is necessary to verify whether
f x f y f x, y for any values x and y within their ranges of definition:
f x x 2 , f y 5 y 2 and f x, y x y 2 x 2 y 4
25 125
3 3 9
we have
25 5 125
f x f y x 2 y 2 x 2 y 2
3 3 9
125
x y 2 x 2 y 4 f x, y , 0.03 x 0.03, 0.15 y 0.15
9
E XY E X E Y
ANSWER: A
15
27. As a consequence of the independence of X and Y
E XY E X E Y E XY E X E Y 0
Therefore
cov X , Y E XY E X E Y 0
ANSWER: A
cov X , Y 0
Therefore
cov X , Y 0
X ,Y 0
SD X SD Y SD X SDY
ANSWER: B
29. We can apply the formula for the expected value of the sum of two functions of X
and Y :
E g1 X g 2 Y E g1 X E g 2 Y
E R E 0.10 X 0.90Y
E 0.10 X E 0.90Y
0.10 E X 0.90 E Y
ANSWER: C
16
30. If a bX is a linear combination of X , c dY is a linear combination of Y , and
R is their sum, i.e., R a bX c dY – where a, b, c and d are constants – the
variance of R , var R , is given by the formula for the variance of linear
combinations of two random variables:
ANSWER: C
31. For the test of the hypothesis that the population mean output per worker per day
(productivity) on Fridays and on days between Monday and Thursday is the same,
against the alternative hypothesis that the population mean output per worker per day
on Fridays is smaller than that on days between Monday and Thursday, i.e. that
productivity drops on Friday, the null and alternative hypotheses are:
H 0 : 1 2 H 1 : 1 2
that is
H 0 : 1 2 0 H 1 : 1 2 0
ANSWER: E
32. For the large sample of workers observed on Fridays ( n1 50) taken from a finite
population ( N 560) without replacement, the sampling ratio is:
n1 50
0.089 0.10 (10%)
N 560
Therefore we do not need to use the finite population correction of the variance.
17
For the large sample of workers observed on days between Monday and Thursday (
( n 2 70 ) taken from a finite population ( N 560) without replacement, the
sampling ratio is:
n2 70
0.125 0.10 (10%)
N 560
N n2
Therefore we need to use the finite population correction of the variance: .
N 1
The two samples are independent. Sampling theory establishes that the sampling
distribution of the difference between the two sample means X 1 X 2 is:
2 2 N n2
( X 1 X 2 ) ~ N 1 2 , 1 2
n1 n2 N 1
The population variances 12 and 22 are known. Therefore the test statistic is given
by
( X 1 X 2 ) ( 1 2 ) 0
12 22 N n2
n1 n2 N 1
ANSWER: A
33. It is known that, for workers observed on Fridays, the sample size is n1 50 , the
finite population size is N 1 N 560 , the sample mean output per worker per day is
X 1 66.6 units, and its population variance is 12 55.5 ; for workers observed on
days between Monday and Thursday, the sample size is n2 70 , the finite population
size is N 2 N 560 , the sample mean output per worker per day is X 2 68.8 units,
and its population variance is 22 58.7 . It is also known that the difference between
population means under the null hypothesis is ( 1 2 ) 0 0 . So the value of the test
statistic is:
2 .2
55.5 58.7 490
50 70 559
18
2 .2
1.11 0.8385710.876565
2.2 2.2
1.6196
1.84506 1.35833
ANSWER: A
34. Since the sampling distribution of the difference between the two sample means
X 1 X 2 is normal:
2 2 N n2
( X 1 X 2 ) ~ N 1 2 , 1 2
n1 n2 N 1
( X 1 X 2 ) ( 1 2 ) 0
Z ~ N 0, 1
12 22 N n2
n1 n2 N 1
ANSWER: D
Z ~ N 0, 1
Since the alternative hypothesis takes the form H 1 : 1 2 0 , the test is one-
tailed (left-hand side tail).
z0.95 1.6449
19
5% 1
95%
Z
1.6449 0
ANSWER: C
36.
5% 1
95%
Z
1.6449 0
Reject Do not
H0 reject H 0
if Z 1.6449 , reject H 0
if Z 1.6449 , do not reject H 0
ANSWER: D
37. Since the test statistic Z 1.6196 1.6449 , we do not eject H 0 : 1 2 at the
5% significance level in favour of the alternative H 1 : 1 2 . There is no
significant evidence (at the 5% level) that the population mean output per worker per
day (productivity) on Fridays ( 1 ) is smaller than that on days between Monday and
Thursday ( 2 ) , i.e. there is no significant evidence that productivity drops on Friday.
ANSWER: C
38. For the test of the hypothesis that the variability of weekly deposits has remained the
same, against the alternative hypothesis that it has changed – ie, for the test of the
hypothesis that the population standard deviation of weekly deposits is equal to
£14,142, against the alternative hypothesis that it is different from £14,142, the null
and alternative hypotheses are:
20
H 0 : 14,142 H 1 : 14,142
H 0 : 2 14,142 2 H 1 : 2 14,142 2 or
H 0 : 2 199,996,164 H 1 : 2 199,996,164
ANSWER: A
39. It is known that the random variable X (weekly deposits) is normally distributed
and that the sample is small ( n 13 ). The object of our inference is the population
variance 2 .
n 1 s 2
02
ANSWER: E
40. It is known that the sample standard deviation of weekly deposits is s 21,512 , the
sample size n 13 , and the population standard deviation under the null hypothesis
0 14,142 . So the value of the test statistic is:
n 1 s 2
13 1 21,512 2
12 462,766,144 5,553,193,728 27.767
2
0 14,142 2
199,996,164 199,996,164
ANSWER: C
41. It is known that the random variable X (weekly deposits) is normally distributed
and that the sample is small ( n 13 ).
ANSWER: E
21
42. The test statistic follows the 2 distribution:
Since the alternative hypothesis takes the form H 1 : 2 199,996,164 , the test is
two-tailed.
0
3.074 28.300
122
ANSWER: C
43.
0
3.074 28.300
122
ANSWER: A
22
44. Since the test statistic 3.074 2 27.767 28.300 , we do not reject
H 0 : 14,142 at the 1% significance level in favour of the alternative
H 1 : 2 14,142 . There is no significant evidence (at the 1% level) that the
population standard deviation of weekly deposits is different from £14,142 – ie, there
is no significant evidence (at the 1% level) that the variability of weekly deposits has
changed.
ANSWER: D
45. For the large sample of car owners in Amsterdam ( n1 2200 ) taken from a finite
population ( N 1 220000 ) without replacement, the sampling ratio is:
n1 2200
0.01 0.10 (10%)
N1 220000
Therefore we do not need to use the finite population correction of the variance.
Sampling theory establishes that the sampling distribution of the sample proportion P1
of hybrid-electric car owners in Amsterdam is:
1 1
P1 ~ N 1 , 1
a
n1
ANSWER: C
46. From the sampling distribution of the sample proportion P1 , the standard error of the
sample proportion P1 is:
1 1 1
SE P1
n1
P1 1 P1
SE P1
n1
23
P1 1 P1 0.057 1 0.057 0.057 0.943
SE P1
0.053751
0.0049
n1 2200 2200 2200
ANSWER: A
47. The 95% confidence interval for the population proportion 1 of hybrid-electric car
owners in Amsterdam is given by
P1 1 P1
1 P1 z0.025 SE P1 P1 z0.025
n1
ANSWER: B
48. From the distribution of the standard normal random variable Z ~ N 0, 1 we have
2.5% 1 2.5%
95%
0 Z
1.9600 1.9600
z 0.025 1.9600
ANSWER: C
49. It is known that the sample proportion of hybrid-electric car owners in Amsterdam is
P1 0.057 (5.7%) , the critical value corresponding to the 95% confidence level
0.0473, 0.0667
24
ANSWER: D
50. The population proportion of hybrid-electric car owners in Amsterdam is some value
between 0.0473 and 0.0667 with a 95% probability. Since this interval does not
include the value 0.08, the confidence interval obtained does not include the
possibility that in Amsterdam the population proportion of hybrid-electric car owners
is 8%, as claimed. There is no evidence (with 95% confidence) in favour of the
claim.
ANSWER: C
51. For the test of the null hypothesis that the population proportion of hybrid-electric car
owners in Vienna ( 2 ) is the same as in Amsterdam ( 1 ), against the alternative
hypothesis that the population proportion of hybrid-electric car owners in Vienna is
smaller than in Amsterdam, the null and alternative hypotheses are:
H0 : 2 1 H1 : 2 1
that is
H0 : 1 2 0 H1 : 1 2 0
ANSWER: C
52. For the large sample of car owners in Amsterdam ( n1 2200 ) taken from a finite
population ( N 1 220000 ) without replacement, the sampling ratio is:
n1 2200
0.01 0.10 (10%)
N1 220000
Therefore we do not need to use the finite population correction of the variance.
For the large sample of car owners in Vienna ( n 2 3100 ) taken from a finite
population ( N 2 314000 ) without replacement, the sampling ratio is:
n2 3100
0.0099 0.10 (10%)
N 2 314000
Therefore we do not need to use the finite population correction of the variance.
Sampling theory establishes that the sampling distribution of the difference between
the two sample proportions P1 P2 is:
25
P1 P2 ~a N 1 2 , 1 1 1 2 1 2
n1 n2
ANSWER: B
53. The population proportions 1 and 2 are unknown. So we use the sample
proportions P1 and P2 , the unbiased point estimators of 1 and 2 , in estimating the
standard error of the difference between the two sample proportions P1 P2 ,
SE P1 P2 .
P1 P2 1 2 0 P1 P2 1 2 0
P1 1 P1 P2 1 P2
SE P1 P2
n1 n2
It is known that, for Amsterdam, the sample proportion of hybrid-electric car owners
is P1 0.057 (5.7%) and the sample size n1 2200 ; for Vienna, the sample
proportion of hybrid-electric car owners is P2 0.043 ( 4.3%) and the sample size
n2 3100 . It is also known that the difference between population proportions under
the null hypothesis is 1 2 0 0 . So the value of the test statistic is:
0.014
0.057 0.943 0.043 0.957
2200 3100
0.014
0.053751 0.041151
2200 3100
0.014 0.014
2.2799
0.000024 0.000013 0.000038
ANSWER: A
26
54. Since the sampling distribution of the difference between the two sample proportions
P1 P2 is normal:
P1 P2 ~a N 1 2 , 1 1 1 2 1 2
n1 n2
P1 P2 1 2 0
Z ~ N 0, 1
P1 1 P1 P2 1 P2
n1 n2
Since the alternative hypothesis takes the form H 1 : 1 2 0 , the test is one-
tailed (right-hand side tail).
z0.05 1.6449
1 5%
95%
0 Z
1.6449
ANSWER: C
55.
1 5%
95%
0 Z
1.6449
Do not Reject
reject H 0 H0
27
if Z 1.6449 , reject H 0
if Z 1.6449 , do not reject H 0
ANSWER: C
ANSWER: A
57. Since the sampling distribution of the difference between the two sample proportions
P1 P2 is normal:
P1 P2 ~a N 1 2 , 1 1 1 2 1 2
n1 n2
P1 P2 1 2 0
Z ~ N 0, 1
P1 1 P1 P2 1 P2
n1 n2
Since the alternative hypothesis takes the form H 1 : 1 2 0 , the test is one-
tailed (right-hand side tail).
z 0.01 2.3263
1
1%
99%
0 Z
2.3263
28
ANSWER: B
58.
1
1%
99%
0 Z
2.3263
Do not Reject
reject H 0 H0
if Z 2.3263 , reject H 0
if Z 2.3263 , do not reject H 0
ANSWER: A
59. Since the test statistic Z 2.2799 2.3263 , we do not reject H 0 : 2 1 at the
1% significance level in favour of the alternative H 1 : 2 1 . There is no
significant evidence (at the 1% level) that the population proportion of hybrid-electric
car owners in Vienna ( 2 ) is smaller than in Amsterdam ( 1 ).
ANSWER: C
60. The evidence against the null hypothesis H 0 : 2 1 and in favour of the
alternative hypothesis H 1 : 2 1 is not very strong: H 0 is rejected at the 5% , but
not at the 1% significance level. So evidence from the study in favour of the claim
that the population proportion of hybrid-electric car owners in Vienna ( 2 ) is smaller
than in Amsterdam ( 1 ) is not very strong: it would be advisable to select larger
samples.
ANSWER: B
29