Scribd
Upload
5 views3 pages

Ariel Compact Data Normality Analysis

This document presents the results of several statistical analyses: 1) A goodness-of-fit test finds that data on the Ariel Compact follows a normal distribution. 2) An ANOVA test finds significant differences between the means of 5 brands, rejecting the null hypothesis that all means are equal. 3) A variance ratio test cannot reject the null hypothesis that the variances of two brands are equal. 4) Brand SE is found to have the highest average image score and a 95% confidence interval is calculated for this mean. 5) A 90% confidence interval is calculated for the population variance based on a sample of 100 observations.

Uploaded by

Aditya Aashish
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
5 views3 pages

Ariel Compact Data Normality Analysis

This document presents the results of several statistical analyses: 1) A goodness-of-fit test finds that data on the Ariel Compact follows a normal distribution. 2) An ANOVA test finds significant differences between the means of 5 brands, rejecting the null hypothesis that all means are equal. 3) A variance ratio test cannot reject the null hypothesis that the variances of two brands are equal. 4) Brand SE is found to have the highest average image score and a 95% confidence interval is calculated for this mean. 5) A 90% confidence interval is calculated for the population variance based on a sample of 100 observations.

Uploaded by

Aditya Aashish
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Aaditaya Aasish(P19059)

Case Study
a)

Bin Frequency

3.2 1 Airel Compact


3.4 0 6

Frequency
3.6 3
4
3.8 5
2
4 1
0 Frequency
4.2 1

4.8
5
3.2
3.4
3.6
3.8

4.2
4.4
4.6

5.2
More
4.4 2
4.6 2 Bin
4.8 1
5 4
5.2 0
More 0

Null: Ariel Compact data follows Normal Distribution.


Assignment: Ariel Compact data does not follow Normal Distribution.
α – Level = 5%.
From Goodness of fit test we find the below table:

Parameter Value
Mean 4.1125
Variance 0.341809
Standard Deviation 0.584645
Chi-Square statistic 43.23777
Degrees of Freedom 2
5% Chi-Square Critical Value 0.10258
KS – Statistic 0.14187
5% KS Critical Value 0.294

Since .102 > 0.05, we cannot reject null at 5% significance level and hence we can assume
Ariel Compact “Image” values is following normal distribution.

b)Null: µ1=µ2=µ3=µ4=µ5

Alternate: Any two of them are not equal

QM Assignement
Aaditaya Aasish(P19059)

Source of Variation SS Df MS F P-value F crit


Rows 6.8504 19 0.360547 1.887215 0.027545 1.725029
Columns 45.1604 4 11.2901 59.09582 1.4E-22 2.492049
Error 14.5196 76 0.191047
Total 66.5304 99
α – level = 0.05 , R=5
FINV(0.05,4,15)=2.4674
F-critical> 2.4674, Hence, we can reject null at 5% significance level.
c)
Null : 𝜎12 = 𝜎22
Alternate: 𝜎12 ≠ 𝜎22
Hence we need to reject null if 𝑠12/𝑠22 > K or 𝑠12/𝑠22 < L
P(Reject null | H0 true) = 0.05
P(𝑠12/𝑠22 > K or 𝑠12/𝑠22 < L | H0 true) = 0.05
P(𝑠12/𝑠22 > K | H0 true) = 2.5%
P((𝑠12 * 𝜎12 /𝑠22 * 𝜎22 )> K*𝜎12/𝜎22 | H0 true) = 2.5%
When H0 true, 𝜎12= 𝜎22 .
P(Fn1-1,n2-1 > K) = 2.5%

P(F19,19 > K) = 2.5%

K = 2.5264.

𝑠12/𝑠22 = 0.326737/0.251868 = 1.2975

𝐾 > 𝑠12/𝑠22 , cannot reject null at 5% significance level.

Similarly,
P(𝑠12/𝑠22 < L | H0 true) = 2.5%
P(Fn1-1,n2-1 < L) = 2.5%

P(F19,19 < L) = 2.5%

L = 0.39582

𝑠12/𝑠22 = 0.326737/0.251868 = 1.2975

𝐿 < 𝑠12/𝑠22, cannot reject null at 5% significance level.

Brand Average Image Score

QM Assignement
Aaditaya Aasish(P19059)

d)Highest average image score is for brand SE. AC 4.16


It has the “Best Image”. AS 4.185
SE 4.585
SB 4.115
R 2.635

e)SE is the best image brand.

N = 20, Sample Mean = 4.585, Sample Standard Deviation S = 0.3232

Confidence Interval = (Sample Mean - tα/2S/√𝑁 , Sample Mean + tα/2S/√𝑁)


(1-α)% = 95% => α% = 5% => α/2% = 2.5% , t0.025,14 = 2.093

Confidence Interval = (4.585 – 2.093*0.3232/√20, 4.585 + 2.093*0.3232/√20) =


(4.4338,4.7362)
f)N=100, 𝜎=0.67
P(a < 𝜎2 < b) = 90% => P(1/b < 1/ 𝜎2 < 1/a) = 90%
P((N-1)S2/b < (N-1)S2/ 𝜎2 < (N-1)S2/a) = 90%
(N-1)S2/ 𝜎2 will follow χ2N−1 distribution, N = 100.
P((N-1)S2/b < χ299 < (N-1)S2/a) = 90%
Using excel we find that
P(χ299 < 77.04633) = 0.05 => (N-1)S2/b = 77.04633 => b = 0.8635
P(χ299 < 123.2252) = 0.95 => (N-1)S2/a = 123.2252 => a = 0.5399
Confidence interval = (a,b) = (0.5399,0.8635)

QM Assignement

You might also like