#Probability & Statistical Data Analysis/ Slides

CHAPTER 3

Descriptive Statistics

[email protected] : 2020/2021 Sem. 2

2
 Measurement of Central Tendency

• A measure of central tendency of a distribution is a

numerical value that describes the central position of
the data or how the data tend to build up in the
center.

• Measurements:
 Mean
 Mode
 Median

3
 Mean
• Mean is the sum of the observations divided by
the number of observations.
• It is the most common measure of central
tendency,
n

x i
Sample mean: x i 1

n
N

x i
Population mean:   i 1

4
 Example

• Data: 13, 18, 13, 14, 13, 16, 14, 21, 13

Calculation:
(13 + 18 + 13 + 14 + 13 + 16 + 14 + 21 + 13) ÷ 9 = 15

• The mean is 15

5
• The mean is unique for every set of data.
• Meaningful for interval and ratio data.
• Can be affected by outliers – rare observations that
are radically different from the rest.

• Example: 3, 4, 6, 4, 7, 3, 6, 5, 1500
• Mean: 170.89

6
 Mean of Grouped Data

• The formula for the mean of grouped data,

h

fx i i
f1 x1  f 2 x2  ....  f h xh
x i 1

n f1  f 2  ....  f h
where,
𝑓 : frequency in a class or frequency of an observed
value.
𝑥 : class midpoint or an observed value.
𝑛 : number of classes or number of observed values.

7
 Example
Find the mean value for the following data:
Number of
children 1 2 3 4 5 6 7
(𝑥 )
Frequency
5 12 8 3 0 0 1
(𝑓 )

Solution:
∑ 𝑓 𝑥 = 5(1) + 12(2) + 8(3) + 3(4) + 0(5)+ 0(6) + 1(7) = 72
h

fx i i
72
x i 1
  2 .5
n 29
8
 Example
Find the mean value for the following data:
Class interval Frequency
41 - 50 7
51 - 60 10
61 - 70 15
71 - 80 2
81 - 90 6
Total 50

9
 Example

Class interval Midpoint, 𝑥 Frequency, 𝑓𝑖 𝑓𝑥

41 - 50 (41 + 50)  2 = 45.5 7 (45.5 x 7) = 318.5
51 - 60 55.5 10 555
61 - 70 65.5 15 982.5
71 - 80 75.5 2 151
81 - 90 85.5 6 513
Total 50 (n) 2520

fx i i
2520
x i 1
  50.4
n 50

10
 Median
• The median is the middle value when the data are
arranged from smallest to largest.
• To find the median, your numbers have to be listed in
an order, so you may have to rewrite your list first.
• For an odd number of observations, the formula for
the place to find the median is
([the number of data points] + 1) ÷ 2

11
 Example

• Data: 13, 18, 13, 14, 13, 16, 14, 21, 13

• Arrange in order: 13, 13, 13, 13, 14, 14, 16, 18, 21

• There are nine numbers in the list, so the middle one

will be the (9 + 1) ÷ 2 = 10 ÷ 2 = 5th number.

• So the median is 14.

12
• For an even number of observations, the median is
the mean of the two middle numbers.

• Example:

2, 3, 3, 5, 6, 7, 8, 9

(5 + 6) ÷ 2 = 5.5 (median)

13
• The median is meaningful for ratio, interval, and
ordinal data.
• The median is not affected by outliers.

• Example: 3, 4, 6, 4, 7, 3, 6, 5, 1500

3, 3, 4, 4, 5, 6, 6, 7, 1500

• Median: 5

14
 Median of Grouped Data
• The median for the grouped data is given by

where,
(Median class is the first class with the value of cumulative frequency equal
at least N/2)
L : lower limit of median class,
N : total number of observations,
cfp : cumulative frequency of the class preceding the median class,
fmed : frequency of the median class,
W : median class size.
15
 Example
Class N÷2 = 40÷2=20
boundary Cumulative
Class Frequency
frequency .: median class = 51-55

41 - 45 40.5 – 45.5 7 7 L  50.5

46 - 50 45.5 – 50.5 10 17 N  40
51 - 55 50.5 – 55.5 15 32 cf p  17
56 - 60 55.5 – 60.5 2 34 W  (50.5  55.5)  5
61 - 65 60.5 – 65.5 6 40 f med  15

Total 40
70

= 51.5
16
 Mode

• The mode is the value that occurs most often.

• If no number is repeated, then there is no mode for
the list.

17
 Example
Case study: On a cold winter day in January, the
temperature for 9 North American cities is recorded in
Fahrenheit as follows:
-8, 0, -3, 4, 12, 0, 5, -1, 0
What is the mode of these temperatures?

Solution:
 Ordering the data from lowest to highest, we get:
-8, -3, -1, 0, 0, 0, 4, 5, 12
 The mode of these temperatures is 0.
18
 Example
Case study: A marathon race was completed by 5
participants. The time taken by each participant is
recorded as follows:
2.7 hr, 8.3 hr, 3.5 hr, 5.1 hr, 4.9 hr
What is the mode of these times given in hours?

Solution:
Ordering the data from least to greatest, we get:
2.7, 3.5, 4.9, 5.1, 8.3
Since each value occurs only once in the data set, there is
no mode for this set of data.
19
 Example
Case study: In a crash test, 11 cars were tested to
determine what impact speed was required to obtain
minimal bumper damage. The collected data as shown
below:
24, 15, 18, 20, 18, 22, 24, 26, 18, 26, 24
Find the mode of the speeds given in miles per hour.

Solution:
Ordering the data from least to greatest, we get:
15, 18, 18, 18, 20, 22, 24, 24, 24, 26, 26
Since both 18 and 24 occur three times, the modes are 18
and 24 miles per hour. This data set is bimodal.
20
 Mode of Grouped Data

• The first step towards finding the mode of the

grouped data is to locate the class interval with the
maximum frequency.
• The class interval corresponding to the maximum
frequency is called the modal class.

21
• The mode of grouped data is calculated using the
formula:

22
 Example
Compute the mode of the test score below.
Score
Frequency
(Class)
41 - 45 1 .: modal class = 26 - 30
36 - 40 5 l  26  0.5  25.5
31 - 35 7 f0 h5
25.5-30.5 26 - 30 16 f1
f1  14
21 - 25 8 f2
f0  7
16 - 20 2
f2  8
Total

14 − 7
𝑀𝑜𝑑𝑒 = 25.5 + 5 = 28.2
2 ∗ 14 − 7 − 8

23
24
Example

25
 Exercise #1
The owner of a shoe shop recorded the sizes of the feet of
all the customers who bought shoes in his shop in one
morning. These sizes are listed below:

8, 7, 4, 5, 9, 13, 10, 8, 8, 7, 6, 5, 3, 11,10, 8, 5, 4, 8, 6

(a)What is the mean of these values?

(b)What is the median of these values?
(c)What is the mode of these values?

26
 Exercise #2
The table below gives the number of accidents each year
at a particular road junction:
2001 2002 2003 2004 2005 2006 2007 2008

4 5 4 52 10 5 3 5

(a) Calculate the mean, median and mode for the values above.
(b) A road safety group want the council to do some improvement
to make this junction safer. Which measure will they use to
argue for this?
(c) The council don't want to spend money on the road junction.
Which measure will they use to argue that safety work is not
necessary?
28
 Exercise #3
You grew fifty baby carrots using special soil. You dig
them up and measure their lengths (to the nearest mm)
and group the results:
Length (mm)
Find the following: Frequency
Class
(a) mean
(b) median 150 – 154 5
(c) mode 155 – 159 2
160 – 164 6
165 – 169 8
170 – 174 9
175 – 179 11
180 – 184 6
185 – 189 3

30
 Exercise #4
Find mean, median and mode corresponding to the
frequency table of samples of students cars and staff cars
obtained from a college.

Age of vehicles Students Staff

1–3 23 30
4– 6 33 47
7–9 63 36
10 – 12 68 30
13 – 15 19 8
16 – 18 10 0
19 – 21 1 0
22 – 24 0 1

32
 Data Profiles

• Percentile
• Quartile

34
 Percentile
In a population or a sample, the P-th percentile is a
value such that at least P percent of the values take on
this value or less and at least (100-P) percent of the
values take on this value or more.

35
Example

36
• Sort the data set so measurements are in order from
lowest to highest,
Y[1], Y[2], …. , Y[N]
• Calculate,
P
i (N )
100

• If i is not an integer, round up to the next highest

integer k and use Y[k] as the percentile estimate.
• If i is an integer, use (Y[i] + Y[i+1]) ÷ 2 as the
percentile estimate.

37
 Example
Given a set of data: 12, 4, 6, 11, 9,15, 20, 18, 25, 30

i) Calculate 80th percentile.

ii) Calculate 68th percentile.

Solution (i): 80th percentile

•Arrange in order:
4 6 9 11 12 15 18 20 25 30

•N = 10, P = 80 ; i = 80 x 10  100 = 8

Y[8] = 20, Y[9] = 25, P80 = (20 + 25)  2 = 22.5

38
 Example

Solution (ii): 68th percentile

N = 10, P = 68

4 6 9 11 12 15 18 20 25 30

i = 68 x 10  100 = 6.8, k = 7

Y[7] = 18, P68 = 18

39
• The process of finding the percentile that
corresponds to a particular value x is:

number of values less than x

percentile of value x  (100)
total number of values

40
 Example
Given a set of data as follows:
12 4 6 11 9 15 20 18 25 30
Find the percentile corresponding to the value,
Y[k] = 15
Solution:
Arrange the data in order,
4 6 9 11 12 15 18 20 25 30
number of values less than 15 5
percentile of value 15  (100)  (100)  50
total number of values 10

.: The value 15 is at the 50th percentile.

41
 Quartile

The 1st, 2nd, and 3d quartiles are the 25th, 50th, and
75th percentiles respectively.

42
 Example
Q1 – 25th percentile
Q2 – 50th percentile (median)
Q3 – 75th percentile

43
 Example

Q1 Q3

44
Example

45
 Exercise #5
0.7901 0.8044 0.8062 0.8073 0.8079 0.8110
0.8126 0.8128 0.8143 0.8150 0.8150 0.8152
0.8152 0.8161 0.8161 0.8163 0.8165 0.8170

(a) Use the 18 sorted (left to right) weights of regular can drinks to
find the percentile corresponding to the given value.
i. 0.8143
ii. 0.8062
(b) Find the indicated percentile and quartile.
i. P80
ii. Q3
iii. P33
iv. Q1
46
48
 Measures of Dispersion

• Measures of dispersion measure how spread out a

set of data is.

• Measurement elements used are Range, variance,

standard deviation.

49
 Range
• The range is the largest number in a set minus the
smallest number.

• Example: 13, 18, 13, 14, 13, 16, 14, 21, 13

The largest value in the list is 21, and the smallest is

13,
so the range is 21 – 13 = 8.

50
 Variance
• A measure of the dispersion of a set of data points
around their mean value.

• It is a mathematical expectation of the average

squared deviations from the mean.

51
  x  x 
n
2
i
• Sample:
s2  i 1

n 1

 x   
2
• Population: i
2  i 1

52
 Standard Deviation
• A statistic used as a measure of the dispersion or
variation in a distribution, equal to the square root of
the arithmetic mean of the squares of the deviations
from the arithmetic mean.

• It is the square root of the variance.

53
 Exercise #6
Adam has been playing golf on the weekends for the past
three years. Recently, he started keeping track of his
recorded scores. His scores for June and July at his favorite
9-hole (par 36) golf course are provided below:

45 49 42 56 41 36 34 38 41 45 40 42 41 39
38 40 39 36 41

Find the Range, Mean, Variance and Standard Deviation

for the above data.
54
 Measures of Shape
• Skewness
• Kurtosis

56
 Skewness
• Occurs when a distribution is not symmetrical about
its mean.
• A distribution is symmetrical when its median, mean,
and mode are equal.
• A positively skewed (skewed to the right) distribution
occurs when the mean exceeds the median.
• A negatively skewed (skewed to the left) distribution
occurs when the mean is less than the median.

57
 Measuring Skewness

• Formula to measure skewness for univariate data x1,

x2, .., xN :

i1 i
N
( x  x ) 3

Skewness 
( N  1) s 3
x = mean

Number of data s = std deviation

points

58
• For normal distribution (symmetric distribution):
 skewness = 0.
• Any symmetric data should have:
 Skewness value near zero.
 Distribution with mean, median and mode fall at
the same point.

59
 Positive/Right Skewed
• Skewness > 0
 The distribution is asymmetrical and points in the positive
direction.
 Example: Test scores of difficult examination where almost
everyone did poorly on it.
 mode < median < mean

60
Positive/Right Skewed

61
 Negative/Left Skewed
• Skewness < 0
 The distribution is asymmetrical and points in the negative
direction.
 Example: Test scores of difficult examination where almost
everyone did good on it.
 mode > median > mean

62
Negative/Left Skewed

63
64
65
 Kurtosis

• Kurtosis is the statistic which describes the degree of

peakedness or flatness of a probability distribution
relative to the benchmark normal distribution.
• In a similar way to the concept of skewness, kurtosis is
a descriptor of the shape of a probability distribution
• Formula to measure kurtosis for univariate data x1, x2,
.., xN:
i1 i
N
Kurtosis = ( x  x ) 4

( N  1) s 4

66
 Excess Kurtosis
• Excess kurtosis is simply kurtosis − 3.
• A normal distribution has kurtosis exactly 3 (excess kurtosis
exactly 0). Any distribution with kurtosis ≈ 3 (excess ≈ 0) is
called mesokurtic.
• A distribution with kurtosis < 3 (excess kurtosis < 0) is called
platykurtic. Compared to a normal distribution, its tails are
shorter and thinner, and often its central peak is lower and
broader.
• A distribution with kurtosis > 3 (excess kurtosis > 0) is called
leptokurtic. Compared to a normal distribution, its tails are
longer and fatter, and often its central peak is higher and
sharper.
67
68
69
 Example

Mira is interested in the elapse time (in minutes) she spends on

riding a tricycle from home at Taman U to School of Computing,
for three weeks (excluding weekends). She obtain the following
data:
19.09, 19.55, 17.89, 17.73, 25.15, 27.27, 25.24, 21.05, 21.65,
20.92, 22.61, 15.71, 22.04, 22.60, 24.25.

Compute and interpret the skewness and kurtosis.

70
 Example - Solution

i1 i
N
( x  x ) 3

Skewness 
( N  1) s 3

∑ (𝑥 − 21.52) −8.245
= = = −0.0183
15 − 1 𝑠 (14)(3.18)

Interpretation: The skewness here is -0.0183. This value implies that the distribution
of the data is slightly skewed to the left or negatively skewed. It is skewed to the left
because the computed value is negative, and is slightly, because the value is close to
zero.

71
 i1 i
N
Kurtosis = ( x  x ) 4

( N  1) s 4
∑ (𝑥 − 21.52) 3086.1
= = = 2.15
15 − 1 𝑠 (14)(3.18)

Interpretation: For the kurtosis, we have 2.15 implying that the

distribution of the data is platykurtic, since the computed value is less
than 3.

72
72