21BBST08P
Sheet 6
Question 1:
The following data represent the profits (in million dollars) for 10 companies in
the industrial sector:
29 28 11 37 33 25 29 32 31 13
(a) Does the sample contain any extreme values? Justify your answer with a
suitable test. Comment on the results
(b) According to your conclusion in part (a), calculate the best central and the
best absolute dispersion measure. Comment on the results
Solution
A- Steps for box plot
Step 1: Put dataset in order (smallest largest)
11 13 25 28 29 29 31 32 33 37
1 1
Step 2: * Location of Q1 = (𝑛 + 1) = (10 + 1) = 2.75
4 4
* Value of Q1 = Start + [ratio ×distance]
= 13 + [0.75] [25-13] = 13 + 9 = 22 million dollars
Comment: The value of Q1 is equal to 22 million dollars, which represents the
profit at 25% distance of the ordered dataset.
3 3
Step 3: * Location of Q3 = (𝑛 + 1) = (10 + 1) = 8.25
4 4
*Value of Q3 = Start + [ratio ×distance]
= 32 + [0.25] [33-32] = 32.25 million dollars
21BBST08P
Comment: The value of Q3 is equal to 32.25 million dollars, which represents
the profit at 75% distance of the ordered dataset.
Step4: IQR = Q3 – Q1 = 32.25 -22 = 10.25 million dollars
Comment: IQR is equal to 10.25 million dollars, which represents the range of
50% distance of the ordered dataset after removing the highest and the lowest
25% of the ordered dataset.
Step5: LB= Q1- 1.5 IQR = 22 – 1.5 (10.25) = 6.625
UB= Q3+ 1.5 IQR = 32.25 + 1.5 (10.25) = 47.625
Comment:
The lower bound is 6.625 and the upper bound is 47.625 and all values between
these two limits so there is no outlier values in this dataset.
B-
- Since there is no outlier, we need to check the skewness first for
determining the best central measure and the best dispersion
measure.
- Skewness Coefficient Rule;
𝑚𝑒𝑎𝑛−𝑚𝑒𝑑𝑖𝑎𝑛
Skewness Coefficient = 3( )
𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
21BBST08P
To get the value of Skewness Coefficient, there is a need to calculate the
values of mean, median and standard deviation as follows;
Mean:
million dollars
Comment:
The mean profit is equal to 26.8 million dollars which represents the value at the
Centre of the data where the majority of observations are around.
Median:
Step 1: Put dataset in order (smallest largest)
11 13 25 28 29 29 31 32 33 37
𝒏 𝒏
Step 2: * Location of the median= & +𝟏
𝟐 𝟐
𝟏𝟎 𝟏𝟎
= & +𝟏
𝟐 𝟐
= 5th & 6th location
* Value of median
𝟐𝟗+𝟐𝟗
= =29 million dollars
𝟐
Comment: The value of median is equal to 29 million dollars, which
represents the profit at 50% distance of the ordered dataset.
Standard deviation:
21BBST08P
S=standard deviation =
million dollars2
Note: we calculated the mean in the previous step million
dollars
11 26.8 -15.8 249.64
13 26.8 -13.8 190.44
25 26.8 -1.8 3.24
28 26.8 1.2 1.44
29 26.8 2.2 4.84
29 26.8 2.2 4.84
31 26.8 4.2 17.64
32 26.8 5.2 27.04
33 26.8 6.2 38.44
37 26.8 10.2 104.04
Total -------- zero 641.6
S=standard deviation = million dollars
Comment:
Standard deviation of the profits is million dollars, which represents the
average distance between the mean ( ) and the values that are around. This
means that the profits in the sample range around the mean ( ) by 8.44 million
dollars.
( -8.44) --------- ( +8.44)
18.36--------------------- 35.24
21BBST08P
𝑚𝑒𝑎𝑛−𝑚𝑒𝑑𝑖𝑎𝑛
****Skewness Coefficient = 3( )
𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
26.8−29
= 3( ) = -0.78
8.44
Since Skewness Coefficient is less than -0.5, so the data is skewed to the left
(Negatively skewed).
Therefore, the best central measure is the median which is equal to
29 million dollars and the best absolute dispersion measure is
interquartile range which is equal to 10.25 million dollars.