2.17 Moments
In this section, we introduce moments of a random variable which serve to describe the shape of
the distribution of a random variable.
Moments about arithmetic mean (central moments)
Definition 2.22 If X is a discrete random variable, the rth moment of X about the mean Hi
denoted by x, is defined by
Br =E(x—w)" = Yo (rH) Pe 7 = 0,152,
a
IfX is a continuous random variable, the rth moment about the mean is defined by
= [ -" foyar
Remarks (i) Since w= E(X), we can write
= E|X- Efe]
(i) If x1, x25... %n are in dual series, i., values of a random variable, with mean , then the rth
moment about jis
—H)',7=0,1,2,.
From the definition of rth moment we have,
b= Gian
Therefore, we get
=Lp= 1
= Ele-wa = Sa Sart
a
SER) —wEp: =EX|— pA
=E(X]—
Therefore, j4y
E(x] =0 (since y= E(x)
=1, 41 =0 for any variable x.
n= Ls -n)p,Random Variables 109
=2¢ (37 2a +42) pi
= Lan Sante Sn
=E(X?)-?
or #2 =E(X*) (E(x)?
7: Ee oe nel It is denoted by 0%. The positive square root of
Bs iLe:, YaTIBOCe, \¢ standard deviation. The moments about mean are also called the
central moments or simply the moments of the distribution. j1), 2 , 3 , pis are called the first four
‘moments of the distribution.
Find the first four moments of the following series about mean: 3, 6, 8, 10, 18.
1 We have x1 = 3, 12 = 6, 13 = 8, 14 = 10, x5 = 18 (given)
346+8+10+18 _ 45
5 5
=a)
= [o—97 +69)? +(8-9)+ (10-97 +(18-9)7]
128
= F16+9+1+1+81] ee110. Probability and Statistics for Science and Engineering
‘The first four moments about the mean are 0, 97.2, 25.6 and 1588.
2.17.1 Moments about an arbitrary number
Definition 2.23 Let X be @ discrete random variable and A be an arbitrary number. The
moment about A of X denoted by 1, is defined as
H.=El(x—-A)’] = Gr —A)’ pr =0,1,2,..
i
If X is a continuous random variable, then 44, is defined as
t= [ (eA Fe)der=0,1,2,...
The moments about an arbitrary number are also called raw moments.
Relation between 1, and
If u denotes the rh moment of a distribution about th
Peo meat ie mean and #/, denotes the rth moment
Hr =H —"Citl, al + "Cau, o(yh)? +. + (I Wir
Substituting r=2,3,4,... we get
a= 1h "Cah C2 (uh)? = 2h — (uh)?
bth —Cunbah +¥Can (yh)? 205)?
=H Subyi, +2 (uh)?
dull 4h) teu () 4c
~All + 646 (4)? —3 448
Remarks O1=¥ 4%»
=Lp=1
ORF a),Random Variables M1
=Yan-4Zy ps
isi fi
=E(X)-A
HH tA
Example — The first four moments of a distributi
s istribution about x = 2 are : poet
Reser yelnes\ol be nae fear cseiral earn x=2 are 1, 2.5, 5.5 and 26 respectively.
Solution We have ph = 1,44, = 2.5.44, =5.5,p4,=26 and A = 2.
Thus, on = 0,12 = Hh — (u\)? = (2.5) — 1? =1.5
Bs = Hy — alg +2004)?
= 5.5 —3(2.5)(1) +2(1)?
=5.5-7.5+2=0
4 = He — Sag + 64h (Hh) ~3 04)"
= 26 —4(5.5)(1) +6(2.5)(1) +3(1)*
=26-22415-3=16
Hy =0,p2 = 1.5, 45 = 0,44 = 16
2.17.2. Moments about origin
Definition 2.24 The rth moment about origin of a random variable X, denoted by ¥; is the
expected value of X", symbolically
n
Epi
= E(x"
for r=0,1,2,..5 when X is discrete, and
= E(x] = fx sewer
when X is continuous.112. Probabitity and Statistics for Science and Engineering
Relation between 41, and V-
If wis the arithmetic mean of X, then the distribution
Vp = pr + City "Copy 2pP + tHE
Putting r= 1,2,3,..., we get
Vi = 1,Va= sn +1? Vs = ust Spon tye
Remark From the definition of moment about arbitrary number A, we have
= Elsa) ] = E (uA) pyr =0,1,2,. :
Taking A =0, we obtain
ts
He ER = ¥ (xi)'p
ie, p= Vp