STATISTICS FORMULAE
MEASURES OF CENTRAL TENDENCY:
1)
Range = Highest Value – Lowest Value
2)
No. Of Class Intervals = 1+ 3.2 log N (Struge’s Rule)
3)
Class Width = Range / No. Of Class Intervals.
OR
Class Intervals = Range / Class Width
4)
Arithmetic Mean = Pr /n, OR Pa r / N
5)
Median =
ven
K
or u m
stVlul–
–
L o w
Where: L = Lower limit of median class, PCF = Prev ious Cumulative
Frequency, F= Frequency of the Median Class, & H= Width of the CI
6)
Mode = 2 = N
.ns.O
Kf.nCs.Is.g
r o v
Where: Where: L = Lower limit of modal class, f
1 = Highest
Frequency in the Distribution, f
0 = Preceding Frequency to the Highest
Frequency in the Distribution, f
2 = Succeeding Frequency to the Highest
Frequency in the Distribution, & H= Width of the CI
7)
Harmonic Mean =
+
P
!
or
"
P
#
$
8)
Geometric Mean =td/4AmcM
P,I5t:
+
or %&'()*+
P.,I5t:
/
PraveenKumar Keskar 2
9)
Quartiles = 01
+an
2
3 or 01
/en
2
3 or ) m
41
5
6
37.8.9
9
L o v
10)
Deciles = 01
&+1
10
3 or 01
<+1
10
3 or
) m
41
5
=
37.8.9
9
L o v
11)
Percentile = 01
+an
nOO
3 or 01
/en
nOO
3 or
) m
41
5
==
37.8.9
9
L o v
MEASURES OF DISPERSION:
12)
Range = Highest Value – Lowest Value
Co-efficient of Range =
TB_px∞Ktk[];xtXtYEUx∞Ktk[];x
TB_px∞Ktk[];xatYEUx∞Ktk[];x
13) Quartile Deviation =
MNOn
K
Co-efficient of Q.D =
MNOn
ONeOn
14)
Mean Deviation =
P|..|Q
+
or
P.P:s:PQ
/
Co-efficient of MD =
RST+tUSVWTXWI+t.6IYtYST+
RST+
PraveenKumar Keskar 3
15)
Standard Deviation = Z
f..[C
+
or Z
P.f..[C
/
OR
Z
Px
2
n
− 1
Px
n
3
K
or Z
Pfx
2
"
− 1
Pfx
"
3
K
Co-efficient of Variation =
a
bc
× 100
PROBABILITY:
16)
Additional Theorem = PfAfBC= PfAC+ PfBC− PfAiBC
17)
Conditional Theorem = Pj
B
A
kl ht
VfminC
VfmC
18)
Baye’s Theorem = Po
A
E
n
kq ht
VfmCoVf
m
r
kC
PVfr
CoVf
m
r
kC
19)
Binomial Distribution:
Probability Mass Function (P.M.F) =
n
uWt
o fsC
b
o ftC
vb
Where,
r = 0, 1, 2 _ _ _ n, p+q = 1
20)
Poison Distribution
P.M.F =
fS
uv
CofYC
!
.!
Where, x = 0, 1, 2, 3 _ _ _ _ _ _ _ _ ∞
PraveenKumar Keskar 4
21) Normal Distribution =
.Y
x
Where m = mean &
y = Standard Deviation
CORRELATION:
22)
Karl Pearson’s Coefficient of Correlation:
r =
+z:{sfz.Cfz|C
}+z:
hut
fz:C
}+z|
hut
fz|C
23)
Spearman’s Rank Correlation:
r = 1-
[
~fz•
C
+f+
snC
] where D = R1 – R2
If any rank is repeated then;
r = 1 –
[
~fz•
atul–lC
+f+
snC
]
C.F. =
YfY
snC
nK
Where, m = No. of times the rank is repeated
REGRESSION:
24)
If X is dependent on Y=
(r -r[) = b xy (? -?c)
b
xy = r
x.
x{
25) If Y is dependent on X =
PraveenKumar Keskar 5
(y -?c) = byx (r -r[)
b
yx = r
x{
x.
b
yx =
+zb|sfzbCfz|C
+fzb
CtstfzbC
b
xy =
+zb|sfzbCfz|C
+fz|
Cstfz|C
r = }?
:{t× ?
{.
TIME SERIES
26)
Method of Moving Averages
27)
Method of Least Squares
a)
Linear Equation:
y = a + bx
Where, a =
P{
+
; b =
P.{
P.
b)
Quadratic Equation:
y = a + bx + cx
2
Normal Equations are:
Σy = na + bΣx + cΣx
2
Σxy = aΣx + bΣx
2 + cΣx
3
Σx
2y = aΣx
2 + bΣx
3 + cΣx
4
INDEX NUMBERS
28)
Unweighted Simple Average:
a)
Unweighted Arithmetic Mean Index Numbers =
P
01 =
z7
+
PraveenKumar Keskar 6
b)
Unweighted Geometric Mean Index Number =
P
01 = Antilog (
z]E_V
+
)
29)
Unweighted Simple Aggregative Index Numbers =
P
01 =
zV
zV=
× 100
30) Weighted Average Relative Index Numbers:
a)
Weighted Arithmetic Mean Index Number =
P
01 =
z?
z?
b)
Weighted Geometric Mean Index Number =
P
01 = Antilog (
zƒ]E_V
z?
)
31)
Weighted Aggregative Index Number =
a)
Laspeyre’s Price Index Number =
P
01 =
zV
M=
zV=M=
× 100
b)
Paasche’s Price Index Number =
P
01 =
zV
M
zV=M
× 100
c)
Marshall Edgeworth’s Price Index Number =
P
01 =
zV
M=atzV M
zV=M=atzV=M
× 100
d)
Fisher’s Price Index Number =
P
01 = }P
OnfLaspeyre′sC× P
Ontfd„„…‰Š†′…Ct