r.r.fi use indicator others 0 IT In PrEIj Xn I In e EE In E - - PDF document
variance and covariance July 23,2020 f Applications of of Expectation Linearity Variance 2 Covariance and correlation 3 Last time Distributions p if Xsl Bernoulli CP x PREM ipso p i f pin pj I Xu PHX Bincngp Linearity of
variance and
covariance
July 23,2020
f Applications of
Linearity
2
Variance 3 Covariance and correlation
Last time
Distributions
p
if Xsl
x Bernoulli CP
PREM
p
ipso
Xu
Bincngp
PHX
I
f pin pj
i
Linearity of
Expectation
ECC
Xi
Cnxn
e E Exif
EWE
1 Applications of Linearity of
Expectation
Example
Hand back assignments
to n students
at random
Expected number of students who get
back theirhw
Xn
the number of students who get back their hw
tn
use indicator
ttsIhnedirn
tbEs gmaE
PrEIj IT In
Xn
I
e
In
EE Ide EE In
c Xn
E
Iie
In
linearity
I
E
I
E a Prodi
a
s
lxprEIEDeoxprcq.ro
achoo't
e
I
n
The expectednumberof students who get their own
home works
in
a class size he
is
Example Throw
me balls
in
n
bins
Be
The expected number of
empty
bins
X number of empty bins
Help Indicators help
Pr
2
ith bin is empty
n m
Etsi
Presist
B
n
Xs
EI linearity
Ist t
n
EEN EEE Ii
EE.CIIS.EE 5senInT.ED
is
E IITs
Eaxprosi ajslxprczisD DXPRET.es
aC 0,12
n
I
m
s
m
For
her and
n
EG
n
E
C EX
h hut
n l 1a
hero
37N 37
tinamou n1
Ie
bins
will
be empty
2J Variance.se
Examle
Random Walk
Particle 4
p
p
p
p
g
r r
a
r B I N
N
J X G
S
4 3
Z
l
I
z
3
4 5
6
a The Particle
moves
in
dimension
a At each
Zime Step Particle moves
to the right or
left
with equal P.robability
Sn
The Position of
the Particle after
n
moves
Xi
the 2th movement
PRED E
pre'T
L
n
Z
Sn
O
X
1
Xze
Xn
Eti
E
n
CCsn
E j Xi
EEN
E xD
IaxPrcai
t.xprexi D
CDxprcxz.se73
AGH R
s 1 12
C 1 Nz
D
what is the expecteddistance from 0
Isnt
we
work SnI
ttiiiiii.axn5
C
Xie
Xn
2 Eg.Efxixi3 Ei
n
E xE
EGxprfdisaJ
lfprcxi
D
Clfprcxz.si
aef
119
1 12
1 12
I
CEXiXj3s E
a bxPr
XI
a Xjsb
m
a betel
pr Xi
a xPrCXj b
1 4
pr Xist PVCxjsD
preexist PrCXj
I
53
sq
GD
prcxis DPrcxjsD fyc yprcxis.BPrfl.jo D
variance
1k
late
1 1,2
1 14
Sn
Definition
For
a
r v
X
with expectationEcosef
the variance of X is defined
to
be
Varys EUx
w T
what does variance measure
The deviation from
mean
value
The square
root
Cx
FarTX
is
the standard deviation
variable X
Theoreme
Var
Ms
EE h J EExFEEEff
proof
Vaux
E
X e5
linearity
E
K2
2 1 V2
ECE
ECD
EM
EE N
2N
r
12
E Ext
v2
Some properties of
Variance
Var
Xtc
Varun
exercise Var
CX
I Varexy
exercise
i
More Examples
Uniform distribution
Xa
a uniform random variable
the set 91 non
Uniformly off
PIKE In
VIELE zu
Var X
ECE
E Ex
9
EEN E.fi n5aa3 aEa
n InaEa
Ea2 PrExsa
a
h
and
NII
aef's TM
In 2
2
44
4 42
s
DR
6
Varcx
h
I 2
Example
Hand back assignments
to n students
at random
Xn
the number of students who get back their hw
pr
Use indicatorsf
I
ft
studenti gets
theirhw
yn
Ans Ii e Iz i In Es Ii
exercise
ECM
Hitt
in
Eis ZEE
Var X
Var X
EEN
E EXT's E EXT
I
EUR3
ECCEIit
ECI
t zig Eui Ij
DE
Ii
l Pr ZED on
Pr Ii Of Ln
cza s2xPrkisDxPIEf on.Tjen
th
t
2
haha
2 E
L
Z heh 1
varCX
s2
Independent Random Variables
Theorem
For independent random variables
X Y
we
have
ECHIT
EIN ELY
independent
Proofof
X andY
EEEYI
Eae.AE gbaPrExsagY b 2
ab PrasadPREYS
aaPrEx a3JxfbEbPrcxsbD
sEEXJrEEYT.D
Theorem
For independent
random variables
y
we have
varcxey
Varchevarly
proof
CExpect'D
Varulay
Elway53 f Exey
E
212
EEx5ezE
EEY7eEEyI
linearityGO.EEig zEqyyypfqy2J
EExI 2EexLEEYJ7EY7
CEEx9 EExt e ENT Eey5 t2 EEXY7ECx3ECyD
Vav.LK eVaVCy
l2
EExyJ ECx3ECyT
If
x
and y
are independent
star
Vary
If
x and y
are
independent then
remember
Varun y
Varun Evan
Var
Atty
Varix
Harty
varCA
t e Dwarcy
Varix
Vary
Example
kn
is the numberof heads
in
n tosses
a biased coin with heads probability P
ithtoss In
Bin en p
A
p
xn Eti
n n
E
Kris EEE Ii3s
E EE Io
E P hip
jet
inn
Is
Varun
s Var
Ishii
EI var
Ii
Var
Ii
C Ii23
EE zig
2
nPhD
ExpresisBeozptiso
PE p p's Pel p
3
covariance and correlation
Definition
The
covariance
X andY
is
CoV XM EEK G LY kDs EExt
E
EEN
p
Remarks
L
If Xgy are independent
CoV X Y
O
Can
we say
if
corex y
so
Roy are independent
No
the converse
is
not
true
2
Cov
Xix
s E Ex XJ
EEXTECD C EXT ECB
ratex
3
Covariance
is
bilinear
For any collection of random variables
A
In
fy
gym and fixed constants
ai
a
san
g
big skim
Then
Cov
aixigo.ES bjYjJs
EaibjCovLxisyj
4
About CoVCR y
1 The sign
Xy
determines how X and Y are
related
2
The
magnitude
Is
hard
to interpret
Definition Correlation
suppose X and Y
are random variables
with
GCN
CY
so The
correlation
corrcx.yjscorcx.yjo.CN
y
Theorem
I
ayebifaso ocorrcx.gs 1
if
a
correngyys l
Corre E
t
Corry
A
s
l
ecgyg.gg gsECxY3
EExIEaD
Ysaxsb
647601
E xcaxebD ECx3EEaxeb
Gex lat
X
64
6caxebjsevaraf.bg
M
aEEXDibEXDaEEXJ.BE
Lal Vara
t
a
a
l ASO