100% found this document useful (1 vote)
125 views28 pagesBivariate Random Variables (
probality distribution
Uploaded by
Swastik koulCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
100% found this document useful (1 vote)
125 views28 pagesBivariate Random Variables (
probality distribution
Uploaded by
Swastik koulCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
BIVARIATE RANDOM
VARIABLES
DISTRIBUTION OF TWO RANDOM VARIABLES
In the last chapters we considered on}
caple space. But we can define more tha
aapk space. For example if we are.
temperature of patients or height and wei
etain region, then these experiments c:
vitthe help of two random variables.
ly one random variable on a
in one random variables on a
interested in pulse rate and
ight of persons belonging to a
an be described. mathematically
Definition. Two Dimensional Random Variable. Let X and Y be two
raiom variables defined on same sample space S. Then the function
UY) that assigns to each element of S a unique point in R?
is called a
‘wodimensional random variable.
EQ, Y) is a two dimensional random variable on sample space $
into ES, then £ = {(X(o), ¥(o)), w € S } is the space of (X,Y).
HCC isan event, then probability of C is defined as
P(C)=PL(X, Y) € C].
‘1-10 Dimensional Diserete Random Variable
"XandY are both discrete random variables, then two dimensional
dor
"Watiable (X, Y) is called a discrete random variable,
‘ition, Joint p.d.f. of Discrete Random Variable. The joint
lly density function of X and Y is defined as
i
PX =x, ¥=y)=/(x,y), where
9S) 0 and GO DD fy =1.
— ee
bh
‘seanned wih CamScanner 394
If X takes values X1,X2s--9% yy and Y takes val ues.
then probability of the event corresponding to X = sy
>|
RAPIDSOL PROBABILITY,
Dp,
. d
given by P(X = %), Y=¥4)= F (joy ;) and the joint Probabif vay
function can be represented by the following table :
Y Ps
oe
. v1 2 | tay
a | fn) | fGp9) | Foy) [rey
x |f£@an) | S222) F(5259n) fils)
ee laa
sees eae
Sm | LCI) | £92) | oon fm) | Hea
Total | fo(y1) fav) fru)
Here P, (x) is obtained by adding all entries of X = 1 fri
values of Y = y,
ie. Py(x,) =PX= x,)
= LOYAL VQ) + oot LR Iq)
= f,(%,) for i= 1, 2, ...,m.
Sometimes. f; (x,) is also denoted by f(x).
Similarly,
Py) =P(V=y,) =
LOI) tF (9)) mt LO)
“AO)=fy 0, ) for j=1,2,.
Since f(x,), = 1, 2,,
alone, therefore it is called mar
Similarly, f, 0), 7=1,2,
m is probability density function a
ginal probability density function®
+ Ms called marginal p.d.f. of ¥-
_di
‘Seanned wih CamScanner ATE RANDOM VARIABLE
av
rk. IEF ¥) is the joint pap
athe
395
serspiliy oF event C is given by
f Fandom Variables x and Y, then,
P(C)= PUK, Y) € c]= Da.
c
saition. Joint Distribution Fy
fi
“ction, Joint distributi i
sn ion function of x
PNP SRYEY= yy F025)
Asa ey
ition. Marginal Distribution Functions, Marginal distribut
eens of X and Y are defined ag Binal distribution
Fy) = PPK Sx,Y=y)
y
and Fy()= D) PX =x xy,
12. Two Dimensional Continuous Random Variables
IfX and Y are continuous ran
dom variables, then variable (x, Y)is
aaled atwo dimensional continuou:
's random variable,
Definition. Joint p.d.f. of Continuous Random Variable. The Joint
probability density function of X and Y is defined as
P(X=x,Y =y)=S(, y), where
O feonzo and i ff sexyyeea=t,
Definition. The Marginal Probability Density Functions. The marginal
MULOFX is defined as
AO = ffandy.
-0
Sinilaly the marginal pd.f. of Y is defined as
FeO = [fenrae.
|
‘Seanned wih CamScanner q
RAPIDSOL PRO!
396 BABILITY THE
ant Distributi ion. Joint distributi
Definition. Joint Distribution Function. Joint distribution function
and Y is defined as K
os
FO,y)=P(X $4, ¥ Sy)= J J fey aay,
Definition, Marginal Distribution Functions. Marginal dit
bu
functions are defined as toy
R= PK <9=PKSHY<™)=f [fanaa
y@
and Fy ()=POS=PR<%YSH= ff Foy)deg,
Fx (2)
a
Remark, x@) = fr (sy) dy,
ay (y)
=——F (x,y) de
Feo) == Fes
#PF(y)
aay”
2. CONDITIONAL DISTRIBUTIONS
Definition. Conditional Probability Density Function. The conditional
probability density function of Y given X = x is defined as
fay @y)
Fx @)
where fxy (t,») =f (Ge, ») is the joint probability density function of
X and Y, and f,(x) is marginal probability density function of X-
Similarly the conditional probability density function of X siv*
Y =yis defined as
and fy sy) =
fx ld=f la)=
= Say)
Fx v@1=fGe|y)= LXE)
xiv @ IY =F ly) %e)
where f,(y) is marginal probability density function of Y-
| —
‘Scanned wih CamScanner y
RANDOM VARIABLES
39;
Independent Random Variables, Two random vari .
v Variable
nd marginal Pat's 7 ,
x@),
yaTe
wht
ae
yee with joint PAL Fv(x,y) a
a sgesid te independent
fi
fev) = Fx Fy0).
ape 1 Let the bivariate probability distibution of x and Y be giy
Bast en
2 4 5g
Jo |
32 32 32 RD
oe 1 1 1
6, 16 8 8 8 :
tf 1 a 2
3232 64 64 ‘ a
Find (/) marginal probabilities of X and Y
(i) PK=1,Y <2) Wii) POX <1)
(») PCY <5) @) PRK <3,Y <5),
Sul () The probability distribution of X and Y is given by
ees eG Px@)
1 2 z 3 8
0 —_— pt japon tae _—
Meas. Gn) a 2 32
ol 1 11 uw
Gg ee gg 16
| eal 1 tog 2 s
32.32 OH 64 64
eee eee orale:
322320COGHHC3s HH | PY) =
b> oe
‘Sanne wih CamScanner oY
Bi
398 RAPIDSOL PROBABILITY THEOR,
‘The marginal probabilities of X are
Py (0) = P(X =0) =z, (= POK= = 2
Py (2)= PX = ad.
‘The marginal probabilities of Y are
Py) =PCY=1
3 3
= P(Y =2)=
3 'y (2) = PCY = 2) ro
Py(3) = PCY =3) = a, Py =P =a) 13,
Py(S) =Py=9)-5, Py (6) = PCY = 6) = s
(ii) PK =1, ¥ S$ 2)=P(X= 1, Y= 1) + P(X= 1, Y=2)
(ii PR $ 1) = PK = 0) + POR = ves =
(iv) PCY s 5)= PCY = 1) + P(Y = 2) + PLY =3)
+P(Y =4)+ P(Y=5)
- ll 13, 6 24 3
3 3 oan 2a
0) PO <3, Y S5)=P(X=0,Y $5)+P(X=1, ¥ <5)
+P(X =2,Y 5)
Feel
t+ J+] t+ 4-4—-4—
2) (i arate)
G alee )
ee eu eae
32°32" 6d 64 6a
4
‘Seanned wih CamScanner JOM VARIABLES
re RA = ere ne 599
o hy te oe probabil ity distribution of two random variables
9H pe given belo *
a
rind () marginal probabilities of X and Y
(i) PX<2¥ 51) Gi) PX <3) (iv) PK <3,y <2)
) conditional distribution of X given Y =
I and conditional
jaribution of Y for given X = 2.
$i The probability distribution of X and Y is given by
7
: 0 1 2 Seep, Ga)
ee ee ee
36 36 36 36 36
|
. 36 36 36 36 36
ee a
36 36 36 36 36
ee
36 36 36 36 36
Oe oll
36 36 36 36
|
‘Sanne wih CamScanner RAPIDSOL PROBABILITy THE,
‘hy
From above table, the marginal probabilities of X are
f(X=1)= 2 seK= a= 2 LK= 3)= ES Kad
36°
The marginal probabilities of Y are
°
PY = n= s, POY = 1)= 3e POY =2)= =, Py =3)= 0
36°
3 _6
ji) P(X <2,Y S1)= raxenysn= sed = 6.1
(ii) PC Se ciager:
: 10,10 _20_5
=P(K=1)+ P(X =2, + = OLS
(ii) P(X <3) = P(X = 1) + PC = 2 ag G6 co:
(iv) P(X $3, ¥ $2)=P(X=1, Y 52) +P(QK=2,Y <2)
+PK=3,Y 59
~(2+d 2) (Edd) ht
36 36 36) (36 36 36) (36 36 °%
pe 28 eld
36 «18°
(v) The conditional probability function of X given Y is
P(X=x,Y=y)
P(X =2x/¥ =y), =
Dee
P(K=1/Y=
P(X=2/Y=1)=
pxK=3/Y=1=2
‘Seanned wih CamScanner VARIATE RANDOM VARIABLES : 401 |
Bl
. 2 I
=1)=PK=4Y=1) 36 2 |
=4|/Y=1)= =sd=t. }
and P(X=4] ) Pen a9 |
36
Hence conditional distribution of X given Y = 1 is
X: 1 2 3 4
i 1 1 1 2
ees
|
Now; conditional distribution of ¥ for X =2 is piven by |
P(K=2,Y=y
POY =y|X=2) = POS2Y=y)
P(X=2) |
P(X
POY =0|X=2) = PORE2Y=0)
P(X
POY =1|X=2)= PR=2V=D) _
In
Pl
PY =2|X=2) = PC
‘Sanned wih CamScanner ON
RAPIDSOL PROBABILITY THEORY
Example 3. The probabilities of a random observation on
402
a bivariate
population (X, Y) is given by the table :
For each observation pair Probabiligy
1
1), 2, 1), G, 3), (4, 3) pt
(2,0), 2, 1), 3), 4, 3) om
G1, 4, 1), , 2), (2, 2), 3,2), (4, 2), (1, 3), (2, 3) 0
Find () P(X =2) (ii) PCY =3) (iit) PCY =2|X=4)
() Examine whether the two events X
independent
() P(K<4,¥ <2),
Sol. From the given information the joint distribution is given below :
= 4 and Y =2 are
Y
fe
x
aa
20 10 10 4
1 1 1
2) ww fa
3 eal
10 10 20 4
(ee ae ee
10 1 20 4
VO) ee Acces: 1
10 10 10
1
PX =2)=-,
@ PR=2) 4
(i) P=3)=3.,
10
(X=4) ‘
i - =4)= PIK=4)(¥=2)) _ 7p _ 2
P(Y =2|X=4)= =M ul
POY ~21x=4)~ ROSA H 2 - 6-2
4
‘Seamed wih CamScanner Ye
givaRiATE RANDOM VARIABLES 408
P(X=0]Y =0) =
wd = P(X=1/Y=0)=
Hence conditional distribution of X given Y = 0 is
ee -1 0 1
PX=x/¥=0):| 0 1 0
Example 5. A two dimensional random variable (X, Y) has a distribution
aiven by f(x, y) “ze x+y), where x and y can assume the integral
values only 0, 1 and 2 Find the conditional distribution of Y forX=2.
Sol. The joint 7 density function is
Sosy) = + (Qx+y), wherex=0,1,2and y=0, 1,2.
Joint probability distribution of X and Y is
Y
ot | RO
x
of/o 2 Ss]
aS
27. 27, 7
Cee pu
2) eee eee a
27 27 27 27
ko! 6 9 2
27 27 27
Conditional distribution of Y given X =x is
Y=y)
ey =y xa) = ROSE ESD,
‘Seanned wih CamScanner 1
406 RAPIDSOL PROBABILITY TH
Ry
P(Y=
PC
When X =0, P(Y=y|X=0)=
P(X=0,Y=0)
P(X=0)
PCY =0|X=0) =
PCY =1|X=0)=
and P(Y =2|X=0)=
P(X=1,Y=y)
a =y|X=1)=
When X= 1, P(Y=y|X=1) are
iz
=0|X=1=22=2
P(Y=0|X=1) a5
2
3
PIY=1|X=1==,
9,
fa 4
Pee ae ata:
When X=2, P(Y=y|X=1)= ee
4
PY =0|x=2)= 4,
¢ | ) b
PYY=1|X=2)= 5
1s”
and P(Y =0|X=2)= £
15
“J
‘Seanned wih CamScanner wa
aRiATE RANDOM VARIABLES
iV! 407
Hence conditional distribution of y given X =x is
|a ola win
a
27 A tan dimanainnal wnndaee ot tt ne ene
‘Seanned wih CamScanner RANDOM VARIAt
avaniATe BLES 409
and marginal distribution of Y is given by
yf ott fp
_{ 4 | 18 | 30
10?) @ G2et |g:
mple 7. A two dimensional random variable (X, Y) has a bivariate
ajgtbution given Dy
Sey)=eBx44y), wherex=0, 1,2 and y=0, 1,2, 3.
Find (i) cso that /(x, y) becomes a p.d.f,
(ii) joint probability distribution
(iii) marginal probability distributions
(iv) conditional probability distribution of X given Y = y.
sol. () We have, f(x,y) = (3x +4 y) forx=0, 1,2 andy=0, 1,2, 3.
* x y cGx+4y) =1
x20 y=0
2
> De c(BxtO+3x4+44+3x4+843x412)=1
x=0
2
2 cD) 4+12xJ=1 > c[24+(24+12)+24+24)]=1
x=0
© We=1 = c=,
108
7 . 1
ii) Putting c= —, we get
(ii) ae
POy)= Te x4 dy), ¥=0,1,2, 9=0 12,3.
+ 10,0)=0,0, =, f(0,2)= 40,3)",
8” 108
a3 oa oul a
10,0 = £0 D= Te L(s2) TAT a
$0, 0)=5 ea =! 4¢,3)-18,
20)=— £2) jog 1) “og 1 Darr
‘Seanned with CamScanner 410
oN
RAPIDSOL PROBABILITY THER,
Hence joint distribution of X and Y is given by
Ne
x
4 8 12 24
9 | ° tos 08 T08 | Tos
Be aed eis
e 108 108 108 108 108
» |s 2 «a | as
108 108 108 108 108
9 21 33 45
4 | ios ios ios ive | !
(iif) Marginal distribution of X is
= 0 1 2
.| 24 | 36 | 48
4s * | joa | ios | ios
Marginal distribution of Y is
ye 0 1 2 3
-| 9 | 2 [33 | 4s
Fro): 108 _| 108 | 108 | 108
(iv) Conditional distribution of X given Y = y is
PK =x] ¥=y)= PXEHYEy)_ Sy) :
P ) P(Y=y)
When Y =0, PO=x1¥=0)= FE)
P(X=0]¥=0)=0,
P(X=1/Y= = 08.
iz
7
108
a
and P(X =2/Y= y= 18
i
‘Seanned wih CamScanner yr
wa
eRAncom VARIABLES an
A =x/Y=1)- 2G)
21, QK=x/Y= I= 2?
nen Y= 1, PC P(Y=1)*
ae
21-18 =4
: wx-0l¥=) 2 I
108
os
=1)= 108 =!
px=1l¥=1) m3
108
10
vee 108 2210)
wd PR=ZIYSD= “oy = 9y-
108
f(x,2)
=2, P(X=x|Y=2)= :
When ¥ ¢ | ) P(Y=2)
» PK=0/Y=2)=
P(X=1|Y=2)=
ad PX =2| y= 2)= one
33
When y = 3, Pot=rIye3)= LE,
a ‘Seanned with CamScanner }
RAPIDSOL PR'
a2 OBABILTY Th,
12 Hy
23-108 - 12 4
POKR=O1Y=3)" "gs a5” 15°
108
pel |
=3)= 108 a
PO TY=3)= 5 i
108
oe
=2]y=3)= 108 =2
POR=21Y=3)= G57 5
108
Hence conditional probability distribution of X given Y is
lo 1 2 3
x
4 8 4
° [sis
1 7 1
SG
2 10 14 2
a 3 Qi s 3
Example 8, Two fair dice are tossed simultaneously. Let X denotes
number on the first die and Y denotes the number on the second di.
Find P(X + Y = 8), P(X + ¥ > 8), P(X = ¥), P(X + =6/Y=4
PK-Y=2).
Sol. PK + =8)=P(X=6, ¥ =2)+P(X=5, Y=3)
+P(K=4, ¥=4) + P(X=3, Y=5)+P(K=
eed
. 363636 "36°36 yaRIATE RANDOM VARIABLE
all
413
ao.
+PXK+Y= 19)
P(X = Y) = P(doublets) = Pex =} Pye
+P =2, y=)
+PK=3,¥=3) 4 pK =
4Y=4)+POXK=5, ves)
+PX=6,¥=6)
ae
36 6
=4)= PAX+Y=6n y=4) P(K=2, y=4)
=6/¥=4)= VAETSON Y=4) _
PR+Y=6/Y=4) Tay 7
Petit
es
eee eee ce
36 36 36 36" 36 36
36
md P(X ~ ¥=2)= P(X =3, ¥=1) + P(X =4, ¥=2)
+P(X=5, ¥=3)+P(K=6,¥=4)
ate a
36 36 36 36 eos
Example 9, Two dice are rolled. Let X denotes the sum on the two faces
‘and Y the absolute value of their difference. Assuming that the dice are
far, find 3
9 MK=yNW=n) Gy PIK=NNW=3)
ii) PAX=y) (iv) PK + ¥=4) NK - Y=2)).
f
‘Since denotes the sum of the two faces and Y the abacatevleof
ti diference, therefore X is distributed over the integers 2, 3,
is distributed over O25:
el
b>
‘Sanned wih CamScanner )
420 RAPIDSOL PROBABILITY THEORY
Example 14. The joint pA. of two dimensional random (X, ¥)is pg
y
ane me, O
(P.U.2010 (9)
Sol. Given joint p.d.f. is
2 en
xt42%, O4) = J [ronda
2 pRIATE RANDOM VARIABLES,
BW)
424
>X
[Using (9)
‘Scanned wih Camscanner ao. RAPIDSOL PROBABILITY THEOR
CF f : Y
“ Bxanple 15. Let two dimensional continuous random variable (x Y)h,
_joint p.d.f. given by ay
_f6x?y, O Y).
() Find P(X <1|¥ <2).
Sol. Given joint pdf. is
6x7y, O
‘Seanned wih CamScanner ]
RAPIDSOL a
424
: PUK <1). (¥<2)) Ry,
a PKSHY <= pry ca)
Example 16. Let p.d.f. of continuous random variable (x, Yyp
xa e ern
O1)
0) P(X<1/Y<1)-
Sol. Given joint p.d.. is
axty®, 0
se9-{
1
: 3
w ro1) ( ) zi
Nay O O<*
