You are on page 1of 3

CHAPTER 9

9-27 The problem statement implies H


0
: p = 0.6, H
1
: p > 0.6 and defines an acceptance region as 80 . 0
500
400
= s p
and rejection region as 80 . 0 > p
a) o=P( p >0.80 | p=0.60) = P
|
|
|
|
|
.
|

\
|

>
500
) 4 . 0 ( 6 . 0
60 . 0 80 . 0
Z
= P(Z>9.13)=1-P(Zs 9.13) ~ 0
b) | = P( p s 0.8 when p=0.75) = P(Z s 2.58)=0.99506

9-39 a) SE Mean from the sample standard deviation 6827 . 0
12
365 . 2
= = =
N
s

b) A one-sided test because the alternative hypothesis is mu > 99.
c) If the null 8253 . 2
12 / 5 . 2
98 039 . 100
0
=

= z
Because ) 8253 . 2 ( u is close to 1, the P-value = ) 8253 . 2 ( 1 u = 0.002 is very small and close to 0. Thus, the
P-value < o = 0.05, and we reject the null hypothesis at the 0.05 level of significance.
d) 95% lower CI of the mean is

o
s
n
z x
05 . 0

s
12
5 . 2
) 645 . 1 ( 039 . 100

s 8518 . 98
e) If the alternative hypothesis is changed to the mu 99, 4397 . 1
12 / 5 . 2
99 039 . 100
0
=

= z
P-value = 15 . 0 ] 9250 . 0 1 [ 2 )] 4397 . 1 ( 1 [ 2 )] ( 1 [ 2
0
= = u = u Z
Because the P-value > o = 0.05 we fail to reject the null hypothesis at the 0.05 level of significance.

9-43 a) 1) The parameter of interest is the true mean battery life in hours, .
2) H
0
: = 40
3) H
1
: > 40
4) z
x
n
0
=

o /

5) Reject H
0
if z
0
> z
o
where o = 0.05 and z
0.05
= 1.65
6) 5 . 40 = x , o = 1.25
26 . 1
10 / 25 . 1
40 5 . 40
0
=

= z
7) Because 1.26 < 1.65 fail to reject H
0
and conclude the battery life is not significantly greater than 40 at o =
0.05.

b) P-value = 1038 . 0 8962 . 0 1 ) 26 . 1 ( 1 = = u
c)
|
|
.
|

\
|
+ u =
10 / 25 . 1
42 40
05 . 0
z |
= u(1.65 + 5.06)
= u(-3.41)
~ 0.000325
d)
( )
( )
, 844 . 0
) 4 (
) 25 . 1 ( ) 29 . 1 65 . 1 (
) 44 40 (
2
2 2
2
2 2
10 . 0 05 . 0
2
2 2
=
+
=

+
=
+
=
o
o
o
| o
z z
z z
n 1 ~ n

e) 95% Confidence Interval

o
s
s
s
85 . 39
10 / ) 25 . 1 ( 65 . 1 5 . 40
/
05 . 0
n z x

The lower bound of the 90 % confidence interval must be greater than 40 to verify that the true mean exceeds 40
hours.

9-53 a) 025 . 0 1 05 . 0 1 s s p then 975 . 0 95 . 0 s s p
b) 05 . 0 025 . 0 s s p
c) 25 . 0 1 4 . 0 1 s s p then 75 . 0 6 . 0 s s p

9-55 a) degrees of freedom = n - 1 = 10 - 1 = 9
b) SE Mean 296 . 0
10
= = =
s
N
s
, then s = 0.9360.
905 . 1
296 . 0
12 564 . 12
0
=

= t
905 . 1
0
= t with df = 10 1 = 9. The P-value falls between two values: 1.833 (for o = 0.05) and 2.262 (for o =
0.025), so 0.05 = 2(0.025) < P-value < 2(0.05) = 0.1. The P-value > o = 0.05 so we fail to reject the null hypothesis
at the 0.05 level of significance.

c) A two-sided test because the alternative hypothesis is mu not = 12.

d) 95% two-sided CI

|
.
|

\
|
+ s s |
.
|

\
|

n
s
t x
n
s
t x
9 , 025 . 0 9 , 025 . 0


2335 . 13 8945 . 11
10
9360 . 0
262 . 2 564 . 12
10
9360 . 0
262 . 2 564 . 12
s s
|
|
.
|

\
|
+ s s
|
|
.
|

\
|


833 . 1 905 . 1
9 , 05 . 0 0
= > = t t we
reject the null hypothesis at the 0.05 level of significance.

f) Reject the null hypothesis that the = 11.5 versus the alternative hypothesis ( 11.5) at the 0.05 level of
significance because the = 11.5 is not include in the 95% two-sided CI on the mean.


9-79
a) In order to use the _
2
statistic in hypothesis testing and confidence interval construction, we need to assume that the
underlying distribution is normal.
1) The parameter of interest is the true standard deviation of titanium percentage, o. However, the solution can be
found by performing a hypothesis test on o
2
.
2) H
0
: o
2
= (0.25)
2

3) H
1
: o
2
= (0.25)
2

4) _
0
2
=
( ) n s 1
2
2
o

5) Reject H
0
if _ _
o 0
2
1 2 1
2
<
/ ,n
where o = 0.05 and _
0 995 50
2
. ,
= 32.36 or _ _
o 0
2
2 1
2
>
, ,n
where o = 0.05 and
_
0 005 50
2
. ,
= 71.42 for n = 51
6) n = 51, s = 0.37
_
0
2
=
( ) ( . )
( . )
.
n s
= =
1 50 0 37
0 25
109 52
2
2
2
2
o

7) Because 109.52 > 71.42 reject H
0
. The standard deviation of titanium percentage is significantly different
from 0.25 at o = 0.01. P-value/2 < 0.005, then P-value < 0.01

b) 95% confidence interval for o:
First find the confidence interval for o
2
:
For o = 0.05 and n = 51, _
o/ , 2 1
2
n
= _
0 025 50
2
. ,
= 71.42 and _
o 1 2 1
2

=
/ ,n
_
0 975 50
2
. ,
= 32.36

36 . 32
) 37 . 0 ( 50
42 . 71
) 37 . 0 ( 50
2
2
2
s so
0.096 s o
2
s 0.2115
Taking the square root of the endpoints of this interval we obtain, 0.31 < o < 0.46

Because 0.25 falls below the lower confidence bound we conclude that the population standard deviation is not
equal to 0.25.


9-85 a) A one-sided test because the alternative hypothesis is p < 0.6

b) The test is based on the normal approximation. It is appropriate because np > 5and n(1 - p) > 5.
c) sample p = 574 . 0
500
287
= =
N
X

( )
1867 . 1
) 4 . 0 )( 6 . 0 ( 500
) 6 . 0 ( 500 287
1
0 0
0
0
=

=
p np
np x
z
P-value = u(-1.1867) = 0.1177

The 95% upper confidence interval is:

6105 . 0
500
) 426 . 0 ( 574 . 0
65 . 1 574 . 0
) 1 (

s
+ s

+ s
p
p
n
p p
z p p
o


d) P-value = 2[1 - u(1.1867)] = 2(1 - 0.8823) = 0.2354


9-89 a)
1) The parameter of interest is the true success rate
2) H
0
: p = 0.78
3) H
1
: p > 0.78
4)
( )
0 0
0
0
1 p np
np x
z

= or
( )
n
p p
p p
z
0 0
0
0
1

= ; Either approach will yield the same conclusion


5) Reject H
0
if z
0
> z
o
. Since the value for o is not given. We assume o = 0.05 and z
o
= z
0.05
= 1.65
6) x = 289 n = 350 83 . 0
350
289
~ = p

( )
06 . 2
) 22 . 0 )( 78 . 0 ( 350
) 78 . 0 ( 350 289
1
0 0
0
0
=

=
p np
np x
z
7) Because 2.06 > 1.65 reject the null hypothesis and conclude the true success rate is greater than 0.78, at o =
0.05.

P-value = 1 - 0.9803 = 0.0197

b) The 95% lower confidence interval:
p
p
p
n
p p
z p
s
s
s

7969 . 0
350
) 17 . 0 ( 83 . 0
65 . 1 83 .
) 1 (

o


Because the hypothesized value is not in the confidence interval (0.78 < 0.7969), reject the null hypothesis.

You might also like