10.

12
Bank 1 Bank 2
4.21 9.66 t-Test: Two-Sample Assuming Equal Variances
5.55 5.9
3.02 8.02 Variable 1 Variable 2
5.13 5.79 Mean 4.286667 7.114667
4.77 8.73 Variance 2.682995 4.335512
2.34 3.82 Observatio 15 15
3.54 8.01 Pooled Var 3.509254
3.2 8.35 Hypothesiz 0
4.5 10.49 df 28
6.1 6.68 t Stat -4.134306
0.38 5.64 P(T<=t) one0.000146
5.12 4.08 t Critical o 1.701131
6.46 6.17 P(T<=t) two0.000293
6.19 9.91 t Critical t 2.048407
3.79 5.47

Critical value = 2.048407

Pooled variance (Sp^2) = 3.509254
Stand error= 0.684033

E= 1.401177

𝛍 1 - 𝛍2 -2.828

-4.229177
-1.426823
Normal ( check normal probability plot)
Assume equal variances

a, Claim: 𝛍1 - 𝛍2 =/ 0

Ho: 𝛍1 - 𝛍2 = 0
H1: 𝛍1 - 𝛍2 =/ 0 (two- tail)

Because T-stat =-4,134 < -2,048 

→ Reject H0

b, 
P-value = 0,0003

P-value <0.05 means that we fail to reject H1. There is evidence of a difference in
mean waiting time between the two branches.

c, The population of waiting time is normally distributed

d, Construct 95% CI for 𝛍1 - 𝛍2

- Critical value: 2.0484 (dùng hàm T.INV.2T)

- Stand error: 1.3164 ( thay số dùng hàm tính)

E = Critical value* Stand error = 2.0484* 0.6840 = 1.4012

→ Construct 95% CI for 𝛍1 - 𝛍2 is ( -4.22918 ; -1.42682 )

10.24
t-Test: Paired Two Sample for Means
Before After D (Before -After)
158 284 -126 Before
189 214 -25 Mean 312.1429
202 101 101 Variance 15513.14
353 227 126 Observations 7
416 290 126 Pearson Correlation 0.295069
426 176 250 Hypothesized Mean Di 0
441 290 151 df 6
t Stat 1.842455
D P(T<=t) one-tail 0.057493
t Critical one-tail 1.94318
Mean 86.14286 P(T<=t) two-tail 0.114986
Standard Error 46.75439 t Critical two-tail 2.446912
Median 126
Mode 126
Standard Deviation 123.7005 Critical value = 2.446912
Sample Variance 15301.81 Stand error= 46.75439
Kurtosis 0.423077
Skewness -0.754195 E= 114.4039
Range 376
Minimum -126 200.5467
Maximum 250 -28.26101
Sum 603
Count 7
Sample for Means a, Claim: 𝛍D > 0

After H0 :  𝛍D ≤ 0
226 H1:   𝛍D  > 0  (right -tail)
4971
7 Excel:   t- Stat = 1.842 < t-Crit = 1.943
→  Fail to reject H0
→ There is not enough evidence for the difference in mean bone marrow micro
density is higher before the stem cell transplant than after.

b, 
P-value = 0.0575 
→ The probability that the t stat for the mean difference in microvessel density is
1.842 is 5.75% if the mean density is not higher before the stem cell transplant
than after the stem cell transplant.

c, Construct 95% CI for 𝛍D

- Critical value: 2.446912 (dùng hàm T.INV.2T)

- Stand error: 46.75439
→ E = Critical value* Stand error = 2.446912 * 46.75439 = 114.4039
→ Construct 95% CI for 𝛍D is (-28.261 ;200.5467 )

d, The population is normally distributed

10.30
Year YES NO a,
2009 0.39 0.61 H0: 𝜋1 <= 𝜋2
2008 0.07 0.93 H1: 𝜋1 > 𝜋2

b, Claim: 𝜋1 > 𝜋2
p ngang = 0.23
H0: 𝜋1 <= 𝜋2
q ngang = 0.77
H1: 𝜋1 > 𝜋2 (right - tailed)
Z stat = 5.376823
Critical value 1.644854 (1: 2009, 2: 2008)

- Test stat
Z stat = 5.376823 (ap dụng công thức)

--> Z stat > Z crit ( 5.376823 > 1.644854 )

--> Reject H0
--> There is sufficient evidence to conclude that the population pr
large online retailers who require three or more clicks to beremov
e-mail list is greater in 2009 than in 2008.

c, Yes, the result in (b) makes it appropriate to claim that the popu
proportion of large online retailers who require three or more clic
removed from an e-mail list is greater in 2009 than in 2008.
 )

de that the population proportion of

more clicks to beremoved from an
.

ate to claim that the population

equire three or more clicks to be
2009 than in 2008.
10.45a
a, Claim: 𝜎^2 𝐴> 𝜎^2 𝐵
Line A Line B
X ngang 8.005 7.997 H0: 𝜎^2 A≤ σ^2 𝐵
S 0.012 0.005 H1: 𝜎^2 𝐴> 𝜎^2 𝐵 (right tail)
n 11 16
- F crit = 2.543719 (dùng hàm F.INV.RT(0.05,10,15))

F crit = 2.543719 - Test stat

F stat = 5.76 (thay số vào công thức 𝑆^21/𝑆^22)
F stat = 5.76
→ 𝐹 𝑠𝑡𝑎𝑡 > F crip
→ Reject H0
→𝐴𝑡 𝑡ℎ𝑒 0.05 𝑙𝑒𝑣𝑒𝑙 𝑜𝑓 𝑠𝑖𝑔𝑛𝑖𝑓𝑖𝑐𝑎𝑛𝑐𝑒, 𝑖𝑠 𝑡ℎ𝑒𝑟𝑒 𝑒𝑣𝑖𝑑𝑒𝑛𝑐𝑒 𝑡ℎ𝑎𝑡 𝑡
𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑖𝑛 𝑙𝑖𝑛𝑒 𝐴 𝑖𝑠 𝑔𝑟𝑒𝑎𝑡𝑒𝑟 𝑡ℎ𝑎𝑛 𝑡ℎ𝑒 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑖𝑛 𝑙𝑖𝑛𝑒 𝐵
 NV.RT(0.05,10,15))

hức 𝑆^21/𝑆^22)

𝑛𝑐𝑒, 𝑖𝑠 𝑡ℎ𝑒𝑟𝑒 𝑒𝑣𝑖𝑑𝑒𝑛𝑐𝑒 𝑡ℎ𝑎𝑡 𝑡ℎ𝑒

𝑎𝑛 𝑡ℎ𝑒 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑖𝑛 𝑙𝑖𝑛𝑒 𝐵
 11.5

Degrees of Mean Square

4 group → c = 4 Source Freedom Sum of Squares (variance)
Among
8 value → n = 8*4 = 32 group c-1=3 SSA = 80*3 = 240 MSA = 80
Within
group n - c =28 SSW = 560 MSW = 560/28 = 20
SST = 240 +560 =
Total n - 1 = 31 800
F

F Stast = 80/20 = 4
 11.1

a, A B C D E Anova: Single Factor

15 16 8 5 12
18 17 7 6 19 SUMMARY
17 21 10 13 18 Groups
19 16 15 11 12 A
19 19 14 9 17 B
20 17 14 10 14 C
Mean 18 17.66667 11.33333 9 15.33333 D
Median 18.5 17 12 9.5 15.5 E

ANOVA
Source of Variation
Between Groups
Within Groups

Total

b,
A B C D E
15 16 8 5 12 Tukey-Kramer Procedure
18 17 7 6 19 Comparison
17 21 10 13 18 A to B
19 16 15 11 12 A to C
19 19 14 9 17 A to D
20 17 14 10 14 A to E
Mean 18 17.66667 11.33333 9 15.33333 B to C
B to D
B to E
c = Factor levels 5 C to D
n 30 C to E
D to E

c, Absolute residuals

A B C D E Anova: Single Factor

3.5 1 4 4.5 3.5
0.5 0 5 3.5 3.5 SUMMARY
1.5 4 2 3.5 2.5 Groups
0.5 1 3 1.5 3.5 A
0.5 2 2 0.5 1.5 B
1.5 0 2 0.5 1.5 C
D
E

ANOVA
Source of Variation
Between Groups
Within Groups

Total
 a,
Single Factor H0: µA = µB = µC = µD = µE
H1: At least one population mean is different.

Count Sum Average Variance F stat > F crip -> Reject H0

6 108 18 3.2
6 106 17.66667 3.866666667 --> there is evidence of a difference in the mean rating o
pens following exposure to five advertisements
6 68 11.33333 11.86666667
6 54 9 9.2
6 92 15.33333 9.466666667

SS df MS F P-value F crit
377.866666666667 4 94.46667 12.56205674 9.7437E-06 2.75871
188 25 7.52

565.866666666667 29

Qa 4.153
Kramer Procedure Num df 5 Den df (n-c) 25
b, With 95% confidence, we can conclude th
Absolutr Difference Critical Range Result
- Advertisements A are different from Advert
0.333333333333332 4.649381960361 Not significantly Different - Advertisements B are different from Advert
6.66666666666667 4.649381960361 Mean significantly Different - Advertisement D is different from Advertise
9 4.649381960361 Mean significantly Different
2.66666666666667 4.649381960361 Not significantly Different
6.33333333333333 4.649381960361 Mean significantly Different
8.66666666666667 4.649381960361 Mean significantly Different
2.33333333333333 4.649381960361 Not significantly Different
2.33333333333333 4.649381960361 Not significantly Different
4 4.649381960361 Not significantly Different
6.33333333333333 4.649381960361 Mean significantly Different

c,
H0: 𝜎^2 1= 𝜎^2 2= 𝜎^2 3
H1: Not all are equal
Single Factor F stat < F crit ( 1.9273 < 2.75871 )
--> Fail to reject H0
--> don’t have sufficient evidence to say that
Count Sum Average Variance advertisements is different
6 8 1.333333 1.366666667
6 8 1.333333 2.266666667 d, The advertisements underselling the pen’s
highest mean ratings, and the advertisements
6 18 3 1.6 characteristics had the lowest mean ratings. T
advertisement that undersells the pen’s charac
advertisements that oversell the pen’s charact
 advertisements is different

d, The advertisements underselling the pen’s

highest mean ratings, and the advertisements
characteristics had the lowest mean ratings. T
6 14 2.333333 2.966666667 advertisement that undersells the pen’s charac
6 16 2.666667 0.966666667 advertisements that oversell the pen’s charact

SS df MS F P-value F crit
14.1333333333334 4 3.533333 1.927272727 0.1371072 2.75871
45.8333333333333 25 1.833333

59.9666666666667 29
= µE
n mean is different.

difference in the mean rating of the

o five advertisements

onfidence, we can conclude that

nts A are different from Advertisements C and D.
nts B are different from Advertisements C and D.
nt D is different from Advertisement E.

^2 2= 𝜎^2 3
equal

1.9273 < 2.75871 )

ct H0
sufficient evidence to say that the variance between the 5
is different

ements underselling the pen’s characteristics had the

atings, and the advertisements overselling the pen’s
had the lowest mean ratings. Therefore, use an
hat undersells the pen’s characteristics and avoid
that oversell the pen’s characteristic
is different

ements underselling the pen’s characteristics had the

atings, and the advertisements overselling the pen’s
had the lowest mean ratings. Therefore, use an
hat undersells the pen’s characteristics and avoid
that oversell the pen’s characteristic
13.4

Shelf Space Weeky Sales

Store (X) (Feet) (Y) ($) SUMMARY OUTPUT
1 5 160
2 5 220 Regression Statistics
3 5 140 Multiple R 0.827001
4 10 190 R Square 0.68393
5 10 240 Adjusted R 0.652323
6 10 260 Standard E 30.80584
7 15 230 Observatio 12
8 15 270
9 15 280 ANOVA
10 20 260 df SS
11 20 290 Regression 1 20535
12 20 310 Residual 10 9490
Total 11 30025

Scatter Plot Coefficients

Standard Error
Intercept 145 21.78302
Weeky Sales (Y) ($)

350
300 Shelf Space 7.4 1.590807
250
200
150
100
50 RESIDUAL OUTPUT
0
4 6 8 10 12 14 16 18 20 22
Observation
Predicted Weeky Sales
Residuals
(Y) ($)
Shelf Space (X) (Feet) 1 182 -22
2 182 38
3 182 -42
4 219 -29
5 219 21
6 219 41
7 256 -26
8 256 14
9 256 24
10 293 -33
11 293 -3
12 293 17
 a, Scatter Plot

b, For each increase in shelf space of an additional foot, predicted

weekly sales are estimated to increase by $7.40

c,
Y ^ = 𝛽0+𝛽1∗𝑥=145+7.4∗8 = 204.2
--> Weekly sales of pet food for stores with 8 feet of shelf space for
pet food is $204,2

MS F Significance F
20535 21.63857 0.000906
949

t Stat P-value Lower 95%Upper 95%

Lower 95.0%
Upper 95.0%
6.656561 5.663E-05 96.4644 193.5356 96.4644 193.5356
4.651727 0.000906 3.855461 10.94454 3.855461 10.94454

204.2

Standard Residuals
-0.749007
1.293739
-1.429923
-0.987328
0.714961
1.395877
-0.88519
0.476641
0.817099
-1.123511
-0.102137
0.578778