STATISTICS FORMULAE By CA Anand V Kaku

No. of class interval × class lengths = Range

LCB = LCL –D/2
UCB=UCL +D/2
Mid-point =LCB+UCB/2 or LCL+UCL/2 1 value value the value
Frequency density of a class interval = frequency of that class interval/corresponding class length
Relative frequency and percentage frequency of a class interval=class frequency/ total frequency
D Decile is given by the Pn Percentile is given by the
n
the Q n quartile is given by the 1 value 1 value
1 value
ARITHMETIC MEAN GEOMETRIC MEAN HARMONIC MEAN MODE
Group Frequency . .

Discrete Series … ….. .

= lower class the
of value
= th at occurs
classthe maximum
GM=
= lower class boundary of the 1 boundary number of
lower
times
boundary of
the Decile∑class i.e. the class the Percentile class i.e. the
= lower class boundary of the Logarithm
Quartile
of Gclass
for ai.e.
set the
of class
containing Decile. class containing Percentile.
∑
median class i.e. the class observations is the
containing AM of the
Quartile.
logarithm of the observations; i.e.
∑ log
log

∑ log
.

Frequency
Distribution …
GM= ….. . X
… ∑ 2
where,
= LCB of the modal class. i.e. the
class containing mode.
= frequency of the modal class
= frequency of the pre – modal
class
= frequency of the post modal class
C = class length of the modal class
 STATISTICS FORMULAE By CA Anand V Kaku
ARITHMETIC MEAN GEOMETRIC MEAN HARMONIC MEAN MODE

Relationship if z = xy, then -

variables GM of z =(GM of x)×(GM of y) ymo=a+bxmo

if z = x/y then
GM of z =(GM of x)/(GM of y)
Weighted Mean -
Weighted A.M
∑ . ∑
∑ ∑ log .
∑
∑

Combined Mean - -
. .

Important points the algebraic sum of deviations of a - - -

set of observations from their AM is
zero ∑ 0

AM is affected due to a change of - - -

origin and/or scale which implies
that if the original variable x is
changed to another variable y by
effecting a change of origin,

Relation among For Given two positive numbers (A.M) x (H.M) = (G.M)2
Averages
AM GM HM The equality sign occurs, as we have already seen, when all the observations are equal.

Mode = 3 Median – 2 Mean

 STATISTICS FORMULAE By CA Anand V Kaku
MEDIAN QUARTILES (Q 1,Q2& Q 3) DECILES (D 1,D2,D 3,……D 9) PERCENTILES (P 1,P2,P3,….,
P99 )

Discrete Series 1
/Unclassified Data Median Size of Q1 quartile is given by the the D 1 Decile is given by the the P 1 Percentile is given by
2
1 1

Distribution 2 . X X
X X

N = total frequency. N = total frequency.

containing median. N = total frequency.
N = total frequency. p= , , , … . , for D 1, D2 p= , , ,…, for P 1,
= less than cumulative frequency p= , , for Q 1, Q2 , Q3 ,D3,…., D 9 respectively P2 ,P3,…,P 99respectively
corresponding to l1. (Pre median respectively = less than cumulative = less than cumulative
class) = less than cumulative frequency corresponding to frequency corresponding to
= less than cumulative frequency frequency corresponding to l1. . (Pre Decile class) l1. (Pre Percentile class)
corresponding to l2. (Post median (Pre Quartile class) = less than cumulative = less than cumulative
class) = less than cumulative frequency corresponding to frequency corresponding to
= being the upper class boundary frequency corresponding to l2. . (Post Decile class) . (Post Percentile class)
of the median class. (Post Quartile class) = being the upper class = being the upper class
C = - = length of the median = being the upper class boundary of the Decile boundary of the Percentile
class. boundary of the Quartile class. class. class.
C = - = length of the Quartile C = - = length of the C = - = length of the
yme= a + b x me class. Decile class. Percentile class.
∑ is minimum if we choose A
as the median.
 STATISTICS FORMULAE By CA Anand V Kaku
DISPERSIONS
Absolute Relative If y = a+bx
RANGE (R) Range= Laregest (L) – Smallest (S) Co efficient of Range = 100 Ry = | | Rx
MEAN DEVIATION (M.D)
M.DA= ∑| | Co efficient of M.D from A = M.Dy = | | M.Dx
about A .D
100
MEAN DEVIATION (M.D)
M.D about Mean = ∑| | Co efficient of M.D from A.M = M.Dy = | | M.Dx
about A.M ( ) .D X
100
MEAN DEVIATION (M.D)
M.D about Median = ∑| | Co efficient of M.D from Median M.Dy = | | M.Dx
about Median .D
= 100
STANDARD DEVIATION (σ) Co efficient of Variation = 100 σy = | | σx
∑ x X
σ
n

∑x
σ X
n
Standard Deviation for Two numbers,
Combined Standard Deviation, σ12=
| | Deviation for First n Natural
Standard
Where d1= , d2=
numbers ,

QUARTILE DEVIATION
(Qd) Qd= Co-efficient of Q.D = 100
Or
.
Co-efficient of Q.D = 100

VARIANCE (σ2) Variance means Square of Standard Deviation

 STATISTICS FORMULAE By CA Anand V Kaku

CORRELATION

KARL PEARSON’S PRODUCT MOMENT SPEARMAN’S RANK CORRELATION COEFFICIENT OF CONCURRENT

CORRELATION COEFFICIENT COEFFICIENT DEVIATIONS

& 6∑
1
1
Where,
∑ where c is concurrent deviation,
&
∑
For tied ranking,
m is one less than number of pairs of
6 ∑ ∑
12 observations
∑ X ∑ 1
1

∑ ∑ ∑

∑ ∑ ∑ ∑

where
| || |
 STATISTICS FORMULAE By CA Anand V Kaku
REGRESSION ANALYSIS
Y depends on X X depends on Y
Simple Regression y= a + byxx x = a + bxyy Intersection
Equation point of these
two lines is ,
Normal Equations b x b y

x b x y b y
Regression Co , ,
b b
efficient
b b
∑ ∑ ∑ ∑ ∑ ∑
b b
∑ ∑ ∑ ∑

b b
where

1
 STATISTICS FORMULAE By CA Anand V Kaku
NAME CONDITION PROBABILITY MASS FUNCTION Notation MEAN VARINACE MODE Remarks
Binomial Trials are independent p+q=1
n x n-x If (n+1)p is integer then
Distribution and each trail has only P(X=x) = Cx p q X ∼B (n,p) μ=np σ2=npq
two outcomes Success Mode, μ0 = (n+1)p-1
& Failure.
If (n+1)p is non-integer then
For x = 0,1,2,..,n Mode,μ0 =Highest Integer in
(n+1)p
Poisson Trials are independent X ∼P (m) μ=m σ2=m If m is Integer Mode,μ0 =m-1 e = 2.71828
Distribution and probability of .
occurrence is very small P(X = x) = If m is non- integer Mode, μ0
in given time.
!
= Highest Integer in m
For x = 0,1,2,..,n

Normal or When distribution is X ∼N (μ,σ2) Mean σ2 μ Mean Deviation =

Gaussian symmetric = 0.8 σ
Distribution Median
/ z= = First Quartile
σ√ Mode = μ- 0.675 σ
=μ
Third Quartile
For ∞ ∞ = μ+ 0.675 σ

Quartile
Deviation=0.675σ

Point of Inflexion
x=μ- σand x= μ+ σ