List of formulas
1. Mean:
(i) Ungrouped data:
𝑆𝑢𝑚 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠
𝑥̅ =
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠
(ii) Discrete Data:
a) Direct Method:
∑ 𝑓𝑥
𝑥̅ =
∑𝑓
b) Short-cut Method/Assumed Mean Method:
∑ 𝑓𝑑
𝑥̅ = 𝐴 + ∑𝑓
, 𝑑 = 𝑥 − 𝐴, 𝐴 is the assumed mean.
(iii) Continuous Data:
a) Direct Method:
∑ 𝑓𝑥
𝑥̅ =
∑𝑓
b) Short-cut Method/Assumed Mean Method:
∑ 𝑓𝑑
𝑥̅ = 𝐴 + ∑𝑓
, 𝑑 = 𝑥 − 𝐴, 𝐴 is the assumed mean.
c) Step-Deviation Method:
∑ 𝑓𝑑′ 𝑑
𝑥̅ = 𝐴 + ( ∑𝑓
) ℎ, 𝑑′ = ℎ , 𝑑 = 𝑥 − 𝐴, 𝐴 is the assumed mean,
ℎ is the height of the interval.
2. Median:
(i) Ungrouped Data:
𝑁+1 𝑡ℎ
a) If the number of observations is odd, then Median is ( ) observation.
2
b) If the number of observations is even, then Median is the average of
𝑁 𝑡ℎ 𝑁 𝑡ℎ
(2) 𝑎𝑛𝑑 ( 2 + 1) observation.
(ii) Discrete Data:
𝑁+1
After finding the cumulative frequency (cf), find the value of ( ). Figure out the
2
𝑁+1
value which is either equal to the value of ( ) or the first bigger value than it in the
2
column of cf. The corresponding value of x will be the median value.
(iii) Continuous data:
𝑁
−𝑐
𝑀𝑒𝑑𝑖𝑎𝑛 = 𝑙 + ( 2 𝑓 ) ℎ,
where, 𝑙 is lower limit of the median class,
𝑐 is the cumulative frequency of the class preceding the median class,
𝑓 is the frequency of the median class,
ℎ is height of the median class.
3. Mode:
(i) Continuous data:
1 0 𝑓 −𝑓
𝑀𝑜𝑑𝑒 = 𝑙 + (2𝑓 −𝑓 ) ℎ,
−𝑓 1 0 2
Where, 𝑙 is lower limit of the modal class,
𝑓1 is the highest frequency,
𝑓0 is the frequency of the class preceding the modal class,
𝑓2 is the frequency of the class succeeding the modal class,
ℎ is height of the modal class.
4. Range:
The difference between the largest and the smallest observations.
5. Quartile Deviation (Q.D.):
(i) Continuous data:
𝑄3 − 𝑄1
𝑄. 𝐷. =
2
𝑁
−𝑐
4
𝑄1 = 𝑙 + ( ) ℎ,
𝑓
where, 𝑙 is lower limit of the first quartile class,
𝑐 is the cumulative frequency of the class preceding the first quartile class,
𝑓 is the frequency of the first quartile class,
ℎ is height of the first quartile class.
3𝑁
−𝑐
𝑄3 = 𝑙 + ( 4 𝑓 ) ℎ,
where, 𝑙 is lower limit of the third quartile class,
𝑐 is the cumulative frequency of the class preceding the third quartile class,
𝑓 is the frequency of the third quartile class,
ℎ is height of the third quartile class.
6. Mean Deviation (M.D.):
(i) Ungrouped data:
(a) Mean deviation from mean:
∑|𝑋𝑖 − 𝑋̅|
𝑀. 𝐷. =
𝑁
(b) Mean deviation from median:
∑|𝑋𝑖 − 𝑀𝑒𝑑𝑖𝑎𝑛|
𝑀. 𝐷. =
𝑁
(c) Mean deviation from mode:
∑|𝑋𝑖 − 𝑀𝑜𝑑𝑒|
𝑀. 𝐷. =
𝑁
𝑀.𝐷.
(d) Coefficient of M.D.= 𝑀𝑒𝑎𝑛 𝑜𝑟 𝑀𝑒𝑑𝑖𝑎𝑛 𝑜𝑟 𝑀𝑜𝑑𝑒
(ii) Grouped data:
(a) Mean deviation from mean:
∑ 𝑓|𝑋𝑖 − 𝑋̅|
𝑀. 𝐷. =
∑𝑓
(b) Mean deviation from median:
∑ 𝑓|𝑋𝑖 − 𝑀𝑒𝑑𝑖𝑎𝑛|
𝑀. 𝐷. =
∑𝑓
(c) Mean deviation from mode:
∑ 𝑓|𝑋𝑖 − 𝑀𝑜𝑑𝑒|
𝑀. 𝐷. =
∑𝑓
𝑀.𝐷.
(d) Coefficient of M.D.= 𝑀𝑒𝑎𝑛 𝑜𝑟 𝑀𝑒𝑑𝑖𝑎𝑛 𝑜𝑟 𝑀𝑜𝑑𝑒
7. Standard Deviation (S.D.):
(i) Ungrouped Data:
∑(𝑋𝑖 − 𝑋̅)2
𝑆. 𝐷. = 𝜎 = √
𝑁
∑(𝑋𝑖 −𝑋̅)2
Variance=𝜎 2 = 𝑁
(ii) Grouped data:
(a) Direct method
∑ 𝑓(𝑋𝑖 − 𝑋̅)2
𝑆. 𝐷. = 𝜎 = √
∑𝑓
(b) Short-cut method
2
∑ 𝑓𝑑 2 ∑ 𝑓𝑑
𝑆. 𝐷. = 𝜎 = √[ −( ) ],𝑑 = 𝑥 − 𝐴
∑𝑓 ∑𝑓
(c) Step-deviation method
2
∑ 𝑓𝑑 ′2 ∑ 𝑓𝑑′ 𝑑
𝑆. 𝐷. = 𝜎 = ℎ√[ −( ) ] , 𝑑′ =
∑𝑓 ∑𝑓 ℎ
𝑑 =𝑥−𝐴
8. Coefficient of Variation (C.O.V):
𝜎
𝐶. 𝑂. 𝑉. = 𝑋̅ × 100
9. Correlation:
(i) Karl Pearson’s Coefficient of Correlation:
a) Direct Method:
𝑛 ∑ 𝑥𝑦 − ∑ 𝑥 ∑ 𝑦
𝑟=
√𝑛 ∑ 𝑥 2 − (∑ 𝑥)2 √𝑛 ∑ 𝑦 2 − (∑ 𝑦)2
b) Short-cut Method:
𝑛 ∑ 𝑑𝑥 𝑑𝑦 − ∑ 𝑑𝑥 ∑ 𝑑𝑦
𝑟= , 𝑑𝑥 = 𝑥𝑖 − 𝐴, 𝑑𝑦 = 𝑦𝑖 − 𝐴
2
√𝑛 ∑ 𝑑𝑥2 − (∑ 𝑑𝑥 )2 √𝑛 ∑ 𝑑𝑦2 − (∑ 𝑑𝑦 )
(ii) Spearman’s Coefficient of Rank Correlation:
a) If there is no tie,
6 ∑ 𝐷2
𝜌=1−
𝑁(𝑁 2 − 1)
where, 𝜌= the coefficient of rank correlation
D= the difference between paired ranks
N= the number of items ranked
b) In case of tie,
𝑚𝑖 (𝑚𝑖2 − 1)
6 [∑ 𝐷2 + ∑ ]
12
𝜌=1−
𝑁(𝑁 2 − 1)
where, 𝜌= the coefficient of rank correlation
D= the difference between paired ranks
m = number of times the value is repeated
N= the number of items ranked
10. Regression equations:
(i) Regression equation of x on y:
(𝑥 − 𝑥̅ ) = 𝑏𝑥𝑦 (𝑦 − 𝑦̅), where,
𝜎
a) 𝑏𝑥𝑦 = 𝑟 (𝜎𝑥 ) , or
𝑦
𝑁 ∑ 𝑥𝑦−∑ 𝑥 ∑ 𝑦
b) 𝑏𝑥𝑦 = , or
𝑁 ∑ 𝑦 2 −(∑ 𝑦)2
𝑁 ∑ 𝑑𝑥 𝑑𝑦 −∑ 𝑑𝑥 ∑ 𝑑𝑦
c) 𝑏𝑥𝑦 = 2 −(∑ 𝑑 ) 2
𝑁 ∑ 𝑑𝑦 𝑦
(ii) Regression equation of y on x:
(𝑦 − 𝑦̅) = 𝑏𝑦𝑥 (𝑥 − 𝑥̅ ), where,
𝜎𝑦
d) 𝑏𝑦𝑥 = 𝑟 (𝜎 ) , or
𝑥
𝑁 ∑ 𝑥𝑦−∑ 𝑥 ∑ 𝑦
e) 𝑏𝑦𝑥 = , or
𝑁 ∑ 𝑥 2 −(∑ 𝑥)2
𝑁 ∑ 𝑑𝑥 𝑑𝑦 −∑ 𝑑𝑥 ∑ 𝑑𝑦
f) 𝑏𝑦𝑥 = 𝑁 ∑ 𝑑𝑥2 −(∑ 𝑑𝑥 )2
11. Baye’s theorem:
𝐴
𝐸𝑖 𝑃(𝐸𝑖 )𝑃 (𝐸 )
𝑖
𝑃( ) = ; 𝑖 = 1,2, … , 𝑛
𝐴 𝐴 𝐴 𝐴
𝑃(𝐸1 )𝑃 (𝐸 ) + 𝑃(𝐸2 )𝑃 (𝐸 ) + ⋯ + 𝑃(𝐸𝑛 )𝑃 (𝐸 )
1 2 𝑛
12. Binomial Probability Function:
𝑓(𝑟) = 𝑛𝐶𝑟 𝑝𝑟 𝑞 𝑛−𝑟 ,
where, 𝑟 is the number of successes,
𝑝 is the probability of a success on one trial,
𝑞 is the probability of failure,
𝑛 is the number of trials
𝑓(𝑟) is the probability of 𝑟 successes in 𝑛 trials.
Expected value or Mean of a Binomial Distribution is
E(𝑥) = 𝑛𝑝
Variance of a Binomial Distribution is
𝑉𝑎𝑟(𝑥) = 𝜎 2 = 𝑛𝑝(1 − 𝑝)
13. Poisson Probability Function:
𝜆𝑥 𝑒 −𝜆
𝑓(𝑥) =
𝑥!
where,𝑓(𝑥) is the probability of 𝑥 occurrences in an interval,
𝜆 is the expected value or mean number of occurrences in an interval.
Expected value or Mean of a Poisson Distribution is
E(𝑥) = 𝜆
Variance of a Binomial Distribution is
𝑉𝑎𝑟(𝑥) = 𝜎 2 = 𝜆
14. Normal Probability Function:
1 𝑥−𝜇 2
1 − ( )
𝑓(𝑥) = 𝜎√2𝜋 𝑒 2 𝜎
where, 𝜇 is mean
𝜎 is standard deviation
