Other Partition Values – Quartiles, Deciles, Percentiles

Meaning values dividing a given set of observations into a number of equal parts
Median Median is also a quartile that divides the set of observations into two equal parts.
Number of equal parts Four (4)
Quartiles Number of Quartiles Three (3)
Denoted by Q1 , Q2 ,Q3
Number of equal parts Ten (10)
Deciles Number of Deciles Nine (9)
Denoted by D1 , D2 ,D3 ,…,D9

Number of equal parts Hundred (100)

Percentiles Number of Percentiles Ninety Nine (99)
Denoted by P1 , P2 ,P3 ,…,P99

(n+1)Pth term,
𝑃𝑡ℎ Quartile 1 2 3
here p= , , ,
4 4 4

How to calculate (n+1)Pth term,

𝑃𝑡ℎ Decile 1 2 3 9
Partition Values here p= , , ,…,
10 10 10 10

(n+1)Pth term,
𝑃 𝑡ℎ
Percentile 1 2 3 99
here p= , , ,…,
100 100 100 100

Mode
Definition Mode is the value that occurs the maximum number of times.
Type of Mode A distribution can be uni-modal, bi-modal or multi-modal
𝑓0 − 𝑓−𝑙
𝑀𝑜𝑑𝑒 = 𝑙1 + [ ] × 𝐶𝑙𝑎𝑠𝑠 𝑙𝑒𝑛𝑔𝑡ℎ
2𝑓0 − 𝑓−𝑙 − 𝑓1
For Frequency Here,
Distribution 𝑓0 = frequency of the modal class
𝑓−1 = frequency of pre modal class
𝑓1 = frequency of the post modal class

Relationship between Mean, Median and Mode

For a symmetric Mean = Median = Mode
distribution
For a moderately Mean – Mode = 3 (Mean – Median)
skewed distribution
Geometric Mean
For a given set of n positive observations, the geometric mean is defined as the nth root of
Definition
the product of the observations
1
Formula
𝐺 = (𝑥1 × 𝑥2 × 𝑥3 … × 𝑥𝑛 )𝑛
1
→ log G = n Σ log x
→ If all observations are constant GM is also constant
Properties
→ 𝐺𝑀 𝑜𝑓 𝑥𝑦 = 𝐺𝑀 𝑜𝑓 𝑥 × 𝐺𝑀 𝑜𝑓 𝑦
𝑥 𝐺𝑀 𝑜𝑓𝑥
→ 𝐺𝑀 𝑜𝑓 =
𝑦 𝐺𝑀 𝑜𝑓 𝑦
Usage GM should be used when observations are in the form of ratios or percentages

Harmonic Mean
For a given set of non-zero observations, harmonic mean is defined as the reciprocal of the
Definition
AM of the reciprocals of the observation
n
Formula H=
Σ( 1⁄x )

→ If all observations are constant HM is also constant

n +n2
Properties
→ Combined HM: x̅ c = n11 n
+ 2
H1 H2

Usage HM is used when observations are in the form of rates like speed, hours per day etc.

Relationship between AM, GM and HM

General AM ≥ GM ≥ HM
When all the observations are same AM = GM = HM
When all the observations are distinct AM > GM > HM

Ideal Measure of Central Tendency

Best Measure – Overall AM
Best Measure for Open End Class Median
Based on all observations AM, GM, HM
Based on 50% values Median
Not affected by Sampling fluctuations Median
Rigidly defined, easy to comprehend AM, Median, GM, HM
No Mathematical Property Mode
Dispersion

Dispersion for a given set of observations may be defined as the amount of deviation of
Definition
the observations, usually, from an appropriate measure of central tendency

These are with units and not useful for comparison of two
Absolute Measures
variables with different units. Example: Range, Mean Deviation,
of Dispersion
Standard Deviation, Quartile Deviation
Types of Dispersion These are unit free measures and useful for comparison of two
Relative Measures of variables with different units. Example: Coefficient of Range,
Dispersion Coefficient of Mean Deviation, Coefficient of variation,
Coefficient of Quartile Deviation

Range
Definition Difference between the largest and smallest of observations.
Formula in case of Range = L – S
Discrete Observation
Formula in case of L = UCB of last class interval
Grouped Frequency S = LCB of first class interval

L-S
Relative Measure Coefficient of Range = × 100
L+S
Change of Origin/ No effect of change of origin but affected by change of scale in the magnitude
Scale (ignore sign)

Mean Deviation
Mean deviation is defined as the arithmetic mean of the absolute deviations of the
Definition
observations from an appropriate measure of central tendency
1
Formula MDA = Σ|x-A|
n
Here A is mean or median as given in question

Mean Deviation about A

Relative Measure Coefficient of Mean Deviation = × 100
A
Change of Origin/ No effect of change of origin but affected by change of scale in the magnitude
Scale (ignore sign)
Standard Deviation
It is defined as the root mean square deviation when the deviations are taken from the
Definition
AM of the observations

Formula Σ(x-x̅ )2 Σx2 2

SDx or σx =√ or√ - (x̅ )
n n
SD
Relative Measure Coefficient of Variation =
AM
× 100

|a-b|
For any two numbers, a and b SD =
Standard Result
2
2
SD of first n natural numbers √(n -1)
12

→ If all the observations are constant, SD is Zero

→ No effect of change of origin but affected by change of scale in the magnitude
Properties of SD (ignore sign)
2 2
n s 2+n2s2 2+n1 d1 +n2 d2
→ Combined SD =√ 1 1
n1+n2

Quartile Deviation
Definition It is defined as the semi-inter quartile range

Q3 -Q1
Formula Qd =
2
Q3 -Q1
Relative Measure Coefficient of Quartile Deviation =
Q3 +Q1
× 100

Ideal Measure of Dispersion

Best Measure – Overall SD
Best Measure for Open End Class QD
Quickest to compute Range
Not based on all observations Range
Difficult to comprehend and less Mathematical Mean Deviation
Rigidly defined, easy to comprehend Mean Deviation, SD, QD
Not affected by Sampling fluctuations QD