Sie sind auf Seite 1von 9

MNGT 212 COURSEWORK 2

Question 1
(a)

x ~ N (500,50)
H : 475
0
H : 475
a
Where = mean daily sales of bread at a particular supermarket
x 500
Test Statistic: z
~ N (0,1)
50
So: Reject H 0 at 5% significant level if:
x 500
1.645
50
Suppose x = 475 from sample data, so:
x 500 475 500

0.5, -0.5 > -1.645


50
50
Hence: Fail to reject H 0 at 5% significant level.
i.e. Based on available evidence we cannot conclude that there has been a decrease
in the daily sales of bread since the competitors marketing campaign.
(b)
(c )
Reject H 0 at 5% significant level if:
x 500
1.645
50
i.e. if x 417.75
500*(1 0.1) 450
Power of test = Prob ( reject H 0 / = 450)
By CLT, if = 450, x ~ N (450,50)
417.75 450
So: Prob ( x 417.75) P( z
) P( z 0.645) 1 P( z 0.645) 1 0.7389 0.2611
50
Hence: Power of test (when = 450) = 0.2611
Sketch out the shape of the power curve marking those points known to
you.

0 .0 0 8

D i s tr ib u t io n o f
t e s t s ta tis ti c i f = 5 0 0

0 .0 0 6
A re a 1- w he n
= 450

D is t r ib u tio n o f
te s t s ta t is tic i f = 4 5 0

0 .0 0 4

0 .0 0 2

4 1 7 .7 5
400

500

600

700

(d )
50
55

x ~ N 500,
and y ~ N 475,

90
35

H : 1 2 0
0
H : 1 2 0
a
Where = mean daily sales of bread at a particular supermarket
Test statistic: z

( x1 x2 ) D0

12 22

n1 n2

~ N (0,1)

So: Reject H 0 at 5% significant level if:


z

( x1 x2 ) D0

12 22

n1 n2

1.645

Suppose that 1 500, 2 475, 1 50 and 2 55 from sample data, so:


( x1 x2 ) D0

12 22

n1 n2

(500 475) 0
502 552

90 35

2.339 1.645

Hence, reject H 0 at 5% significant level.


i.e. Based on available evidence we can conclude that there has been a decrease
in the daily sales of bread since the competitors marketing campaign.
Question 2

(a ) Independent samples assuming Normal distribution or CLT conditions holding


(b)
3.59 3.27 4.6 3.74 3.78 4.03 3.17 3.81 3.93 3.7 3.29 3.43 .1 3.74 3.77
x1
3.730
15
s1

1 N
( xi x )2
N 1 i 1
(3.59 3.73) 2 (3.27 3.73) 2 (4.6 3.73) 2 (3.74 3.73) 2 (3.78 3.73) 2

1
* (4.03 3.73) 2 (3.17 3.73) 2 (3.81 3.73) 2 (3.93 3.73) 2 (3.7 3.73)2
15 1
2
2
2
2
2
(3.29 3.73) (3.43 3.73) (4.1 3.73) (3.74 3.73) (3.77 3.73)
0.365
3.51 3.34 4.23 3.34 3.99 4.1 2.87 3.49 3.25 2.86 3.01 2.93 3.66 3.11 3.21
x2
3.393
15

s2

1 N
( xi x ) 2

N 1 i 1
3.51- 3.393 2 (3.34 - 3.393) 2 (4.23 - 3.393) 2 (3.34 - 3.393)2 (3.99 - 3.393)2

1
* (4.1- 3.393) 2 (2.87 - 3.393) 2 (3.49 - 3.393) 2 (3.25 - 3.393) 2 (2.86 - 3.393) 2
15 1
2
2
2
2
2
(3.01- 3.393) (2.93 - 3.393) (3.66 - 3.393) (3.11- 3.393) (3.21- 3.393)
0.440
H 0 : 1 2 0

H a : 1 2 0
where :
1 = mean maximum force experienced whilst bowling for each bowler at the start of the year

2 mean maximum force experienced whilst bowling for each bowler at the end of the year
( x x ) D0
z 1 2
~ N (0,1)
s12 s2 2
So: Reject H 0 at 5% significant level if:
( x1 x2 ) D0
1.645
s12 s2 2
Suppose x1 3.73 and x 2 3.393 from the sample data so:
( x1 x2 ) D0 3.73 3.393 0

0.419, 0.419 1.645


s12 s2 2
0.365 0.440
Hence: Fail to reject H 0 at 10% significant level.
i.e: Based on the evidence we can not conclude that the forces have decreased over the year.

( a) Independent
(b)
3.59 3.27 4.6 3.74 3.78 4.03 3.17 3.81 3.93 3.7 3.29 3.43 .1 3.74 3.77
x1
3.730
15
1 N
( xi x )2
N 1 i 1

s1

(3.59 3.73) 2 (3.27 3.73) 2 (4.6 3.73) 2 (3.74 3.73) 2 (3.78 3.73)2

1
* (4.03 3.73) 2 (3.17 3.73) 2 (3.81 3.73) 2 (3.93 3.73) 2 (3.7 3.73)2
15 1
2
2
2
2
2
(3.29 3.73) (3.43 3.73) (4.1 3.73) (3.74 3.73) (3.77 3.73)
0.365
3.51 3.34 4.23 3.34 3.99 4.1 2.87 3.49 3.25 2.86 3.01 2.93 3.66 3.11 3.21
x2
3.393
15

1 N
( xi x )2
N 1 i 1

s2

3.51- 3.393 2 (3.34 - 3.393) 2 (4.23 - 3.393) 2 (3.34 - 3.393) 2 (3.99 - 3.393) 2

1
* (4.1- 3.393) 2 (2.87 - 3.393) 2 (3.49 - 3.393) 2 (3.25 - 3.393) 2 (2.86 - 3.393) 2
15 1
2
2
2
2
2
(3.01- 3.393) (2.93 - 3.393) (3.66 - 3.393) (3.11- 3.393) (3.21- 3.393)
0.440
H 0 : 1 2 0

H a : 1 2 0
where :
1 = mean maximum force experienced whilst bowling for each bowler at the start of the year

2 mean maximum force experienced whilst bowling for each bowler at the end of the year
( x x ) D0
t 1 2
~ tdf
s12 s22

n1 n2
2

s12 s22
0.3652 0.4402

n1 n2
15
15

df

2
2
2
2
1 0.3652
1 0.440
1 s12
1 s22

15 1 15
15 1 15
n1 1 n1
n2 1 n2
So: Reject H 0 at 5% significant level if:
( x1 x2 ) D0
s12 s22

n1 n2

27.076 27

t27,0.05 1.703

Suppose x1 3.73, x 2 3.393, s1 = 0.365 and s 2 = 0.440 from the sample data so:
( x1 x2 ) D0
2
1

2
2

s
s

n1 n2

3.73 3.393 0
0.3652 0.440 2

15
15

2.281 1.703

Hence: Reject H 0 at 5% significant level.


i.e: Based on the evidence we can conclude that the forces have decreased over the year.

(c)

Group Statistics
End
Start

Mean

Std. Deviation

Std. Error Mean

1.00

15

3.7300

.36487

.09421

2.00

15

3.3933

.43993

.11359

Independent Samples Test


Levene's Test for

t-test for Equality of Means

Equality of Variances
F

Sig.

df

Sig. (2-

Mean

Std. Error

95% Confidence

tailed)

Difference

Difference

Interval of the
Difference
Lower

Equal

1.016

.322

Upper

2.281

28

.030

.33667

.14757

.03438

.63896

2.281

27.074

.031

.33667

.14757

.03391

.63942

variances
assumed
Start

Equal
variances
not
assumed

p-values is 0.031/2 < 0.05, hence Reject H 0 at 5% significant level.


i.e.: Based on the evidence, we can conclude that the forces have decreased
over the year.
(d)

Ranks
End

Start

Mean Rank

Sum of Ranks

1.00

15

18.97

284.50

2.00

15

12.03

180.50

Total

30

Test Statisticsa
Start
Mann-Whitney U

60.500

Wilcoxon W

180.500

-2.158

Asymp. Sig. (2-tailed)

.031
.029b

Exact Sig. [2*(1-tailed Sig.)]


Exact Sig. (2-tailed)

.030

Exact Sig. (1-tailed)

.015

Point Probability

.001

a. Grouping Variable: End


b. Not corrected for ties.

Ranks
N

End - Start

Mean Rank

Sum of Ranks

Negative Ranks

30a

15.50

465.00

Positive Ranks

0b

.00

.00

Ties

0c

Total

30

a. End < Start


b. End > Start
c. End = Start

Test Statisticsa
End Start
Z

-4.782b

Asymp. Sig. (2-tailed)

.000

Exact Sig. (2-tailed)

.000

Exact Sig. (1-tailed)

.000

Point Probability

.000

a. Wilcoxon Signed Ranks Test


b. Based on positive ranks.

Ranks
N
VAR00002 - VAR00001

Mean Rank

Sum of Ranks

Negative Ranks

12a

9.42

113.00

Positive Ranks

3b

2.33

7.00

Ties

0c

Total

15

a. VAR00002 < VAR00001


b. VAR00002 > VAR00001
c. VAR00002 = VAR00001

Test Statisticsa
VAR00002 VAR00001
Z

-3.011b

Asymp. Sig. (2-tailed)

.003

Exact Sig. (2-tailed)

.001

Exact Sig. (1-tailed)

.001

Point Probability

.000

a. Wilcoxon Signed Ranks Test


b. Based on positive ranks.

The Wilcoxon signed-rank test is a non-parametric statistical hypothesis test


used when comparing two related samples, matched samples, or repeated
measurements on a single sample to assess whether their population mean
ranks differ (i.e. it is a paired difference test). It can be used as an alternative to
the paired Student's t-test, t-test for matched pairs, or the t-test for dependent
samples when the population cannot be assumed to be normally distributed.
Choose: Wilcoxon signed-rank test.
As sample data does not satisfy the assumptions in (a) (neither from Normal
distribution nor CLT holds), so we use non-parametric tools instead. As it requires
comparing 2 related samples (repeated measurements on a single sample to
assess whether their population mean ranks differ), we choose Wilcoxon signedrank test to test the sample.

H0: the forces have not decreased over the year, 1 - 2 0


Ha: the forces have decreased over the year, 1 - 2 > 0
Reject H0 at 5% significant level if p0.05
As 0.001 < 0.005, reject H0 at 5% significant level.
i.e.: Based on the evidence we can conclude that forces have decreased over the
year.

(e) We have the same conclusion from 3 tools that we can conclude that forces
have decreased over the year.

Das könnte Ihnen auch gefallen