Course No: MTH F113: Probability and Statistics

BITS Pilani
Pilani Campus
Course No: MTH F113
Probability and Statistics

Joint Distributions
Sumanta Pasari
BITS Pilani
Pilani Campus sumanta.pasari@pilani.bits-pilani.ac.in
Let’s Watch…
https://www.youtube.com/watch?v=Qfs_mvpA-vQ&t=112s
3 BITS Pilani, Pilani Campus

Step-by-Step
1. Data Analysis
• Collection, classification, interpretation and analysis of data
• Visualization using charts and graphs
• Measurement of central tendency and dispersion
2. Probability
• Probability axioms and rules
• Important probability distributions
• Joint distributions; moments, correlation
3. Statistics
• Sample statistics
any change?
• Parameter estimation
• Testing of hypothesis

What is Probability?

Discrete Distributions
Gambler’s fallacy…
Poop by a bird…
Expectation and Moments

Missed flight…
Continuous Distribution

Joint Distributions
Single Random Variable (Univariate)
Discrete Continuous
Two Dimensional Random variables (Bivariate)
Discrete Continuous
Joint Distributions
 When two or more random variables are involved in a random experiment
 Toss a coin twice; S = {HH, HT, TH, TT};
• Let X and Y be random variables associated with the first and second tossing. If ‘head’
appears, then associated RV is 1, otherwise 0. Events are {(X=0, Y=0), (X=0, Y=1), (X=1,
Y=0), (X=1, Y=1)} . Can you find out what is P(X=i, Y=j), i=0,1, j= 0,1?
• Let X denote the number of ‘heads’ and Y denote the number of ‘tails’. Notice that
X+Y=2. Then possible events are {(X=0, Y=2), (X=1, Y=1), (X=2, Y=0)} . Can you find out
what is P(X=1, Y=1)? Note that in this example, events like (X=0, Y=0), (X=0, Y=1), (X=1,
Y=0), (X=1, Y=2), (X=2, Y=1), (X=2, Y=2) are impossible events, as X +Y =2.
 Two balls are selected at random from a box that contains 3 blue balls, 2 red balls and 3
green balls. If X is the number of blue balls and Y is the number of red balls selected, find
P(X=1, Y=1), P(X+Y≤1).
 Similarly, let X and Y be random variables associated with the height and weight of a
group of students. Find P(150≤X≤180, 50≤Y≤80).

Joint Distributions
Bivariate distribution occurs when we observe two

nondeterministic quantities, one followed by another.
For example : Record the atmospheric temperature

T in Celsius followed by atmospheric pressure P in
pounds per square foot at a random place and time.
This gives two dimensional r.v. (T, P).
n-dimensional random variable  we observe n
nondeterministic quantities in sequence.
Discrete Joint Density
• Let X and Y be discrete RV associated with a random experiment. Then the

ordered pair (X,Y) is called two-dimensional discrete RV.
• The joint pdf for (X,Y) is f XY  x, y   P  X  x and Y  y 
Necessary and sufficient condition for a function to be a discrete joint density:

(i) f XY ( x, y )  0   x, y   2
(ii)  f
all x all y
XY ( x, y )  1
• For any region A in the xy plane, the probability of A is defined as

P( A)   f XY ( x, y)   x, y   A

Marginal Density of Discrete RVs
• From a known joint density f(x,y) for discrete RVs (X,Y), it is easy to find out their
respective density functions.
(i) The marginal density for X , denoted by f X  x  , is given by
f X  x   P  X  x    f XY ( x, y )
all y
(ii) The marginal density for Y , denoted by fY  y  , is given by

fY  y   P Y  y    f XY ( x, y )
all x
Ex.5.1. For a discrete random variable  X , Y  , P  X  i, Y  j   0, i, j  1, 0,1

1 1 1
except P  X  0, Y  1  , P  X  1, Y  1  , P  X  1, Y  1  .
3 3 3
(i) Find the joint density for  X , Y  and (ii) marginal densities for X and Y .

Examples: Discrete Joint Distribution
Sol (Ex.5.1). We represent pdf and marginal densities in a tabular form.

X=x -1 0 1 marginal Y 1 3; x  0

f X  x   2 3; x  1
Y=y f(y)
-1 0 0 1/3 1/3  0; e.w

0 0 0 0 0
1 0 1/3 1/3 2/3 1 3; y  1

marginal X 0 1/3 2/3 Total Sum fY  y    2 3; y  1
 0; e.w
f(x) 1.0 
x2  y
HW.5.1. For  X , Y  , f XY  x, y   P  X  x, Y  y   ; x  0,1, 2,3; y  0,1.
32
Find the marginal densities f X  x  and fY  y  .
x2  y xy
HW.5.2. Verify whether (a) f XY  x, y   ; x  1, 2; y  0,1 (b) f XY  x, y   ; x, y  0,1
16 12
can define discrte pdfs.
Examples: Discrete Joint Distribution
Ex.5.2. Two balls are selected at random from a box that contains 3 blue balls, 2 red
balls and 3 green balls. If X is the number of blue balls and Y is the number of red balls
selected, find pdf for (X,Y) and marginal pdfs. What is P(X+Y≤1)?
X=x 0 1 2 marginal Y
Y=y f(y) Formula used here:
0 3/28 9/28 3/28 15/28  3 2  3 
1 6/28 6/28 0 12/28    
f  x, y        
x y 2 - x - y
2 1/28 0 0 1/28 8
marginal X 10/28 15/28 3/28 Total Sum  
 2
f(x) 1.0
First we find out the events  x, y  such that x+y  1

P  X  Y  1  f  0, 0   f  0,1  f (1, 0)  3 28  6 28  9 28  9 14

Joint Density for Continuous RVs
• Let X and Y be continuous RV associated with a random experiment. Then

the ordered pair (X,Y) is called two-dimensional continuous RV.
Necessary and sufficient condition for f(x,y) to be a continuous joint density
(i) f XY ( x, y )  0,   x, y   2
 
(ii) 
 
f XY ( x, y )dydx  1
b d
The probability is computed as (i) P  a  X  b, c  Y  d     f XY ( x, y )dydx
y y a c
d (ii) D  2
, P( A)   f ( x, y )dA
D D D
c x
x
a b 15 BITS Pilani, Pilani Campus
Joint Density for Continuous RVs
Note: In one dimensional continuous case, the probabilities
correspond to areas under density curve while in the case of
2-D, they corresponds to volumes under density surfaces.
Z
z  f XY ( x, y )
Probability of
The event R
Y (in XY-plane)
is this volume.
R
X
P[ R]   f XY ( x, y )dA
R 16 BITS Pilani, Pilani Campus
Marginal Density of Continuous RVs
• From a known joint density f(x,y) for continuous RVs (X,Y), it is easy to find out
their respective density functions.
(i) The marginal density for X , denoted by f X  x  , is given by


fX  x   f XY ( x, y )dy

(ii) The marginal density for Y , denoted by fY  y  , is given by


fY  y    f XY ( x, y )dx

Remarks:
(i) The marginal densities are also density functions (Verify!)
(ii) Thus, f X  x  and fY  y  will follow the properties of density functions
For n-dimensional RV
• The concept can be extended in an analogous

manner to n dimensions.
• The joint density fX1…Xn(x1,…,xn) is a function of n
random variables.
• Replicate the definition.
18
BITS Pilani, Pilani Campus
Independent Random Variables
Let  X , Y  be a two-dimensional RV with joint density f XY

and marginal densities f X , fY , respectively.
X and Y are independent if and only if

f XY  x, y   f X  x  fY  y  for all x and y

Independent Random Variables
In general for n RVs,

X 1 ,..., X n are indep iff
n
f X1 ... X n ( x1 ,..., xn )   f X i ( xi )
i 1
Here f X1 ... X n ( x1 ,..., xn ) is the joint density

and are f X i ( xi ) the marginal densities of X i .
Exercise 5.1 (From Book)
3. The joint density for (X,Y) is given by:

1
f XY ( x, y )  2 , x  1,2,3,..n
n
y  1,2,3,..n
and f XY ( x, y )  0 elsewhere.
(a) Verify that fXY(x,y) satisfies the
conditions necessary to be a density.
Exercise 5.1 (Q. 3)
(b) Find the marginal densities for X and Y

(c ) Are X and Y independent?
Sol:
1
for n  0, f XY ( x, y )  2  0,  x  1,..., n
n
and y  1,..., n; and equals 0 elsewhere.
also,
n n n
1 1

x 1 y 1 n
2
 
x 1 n
2
n 1
Exercise 5.1 (Q. 3)
n
1
(b) f X ( x)   2  1/ n, x  1, 2,...n
y 1 n
0 e.w.
n
1
fY ( y )   2  1/ n, y  1, 2,...n
x 1 n
0 e.w.
(c) The random variables X, Y are independent:

Exercise 5.1 (Q. 3)
1 1
f X ( x ) fY ( y ) 
n n
1
 2  f XY ( x , y )
n
for 1  x  n,1  y  n while
f XY ( x , y )  0  f X ( x ) fY ( y )
elsewhere.
Exercise 5.1 (Q. 4)
4. The joint density for (X,Y) is given by

2
f XY ( x, y )  ,1  y  x  n
n(n  1)
x, y integers n is a positive integer
(a)Verify that fXY(x,y) satisfies the conditions
necessary to be a density.
(b) Find the marginal densities for X and Y
Exercise 5.1 (Q. 4)
(c ) Are X and Y independent?

(d) Assume that n=5, use the density
to find
P[X  3 and Y  2]
Find also
P[X  3] and P[Y  2]
Exercise 5.1 (Q. 4)
Sol: for 1  y  x  n
2
Since f ( x, y )   0,
n( n  1)
as n is a +ve integer
and f ( x, y )  0 elsewhere.
n n
2
also, 
all ( x , y )
f ( x, y )   
y 1 x  y n( n  1)
n
2

n( n  1)
 (n  y  1)
y 1
Exercise 5.1 (Q. 4)
2
 {n  (n  1)  (n  2)  ...  1}  1
n(n  1)
x
b) f X ( x )   2
n ( n 1)
y 1
2x
 , x  1, 2,...n
n( n  1)
n
2( n  y  1)
fY ( y )  
x y
2
n ( n 1) 
n( n  1)
, y  1, 2,..n

Exercise 5.1 (Q. 4)
c) X and Y are not independen t
2 1 2(n  2  1)
f X (1) fY (2) 
n(n  1) n(n  1)
 0  f XY (1,2)

Exercise 5.1 (Q. 4)
d) when n  5,
2 3
2
P[X  3 and Y  2]  
y 1 x  y 5(6)
2 2
 
30 y1
(3  y  1)  1 / 3
3
2x 2 (2  4  6)
P[X  3]     2/5
x 1 5(6) 30 2
2 2
 
30 y1
(3  y  1)  1 / 3

Exercise 5.1 (Q. 4)
P[Y  2]
2
2(5  y  1) 2
  (5  4)  3 / 5
y 1 5(6) 30

Exercise 5.1 (Q. 9)
(9)An engineer is studying early morning

traffic pattern at a particular intersection.
The observation period begins at 5.30 a.m.
Let X denote the time of arrival of the first
vehicle from the north – south direction.
Let Y denote the first arrival time from the

Exercise 5.1 (Q. 9)
east – west direction. Time is measured in

fractions of an hour after 5.30 a.m. Assume
that density for (X,Y) is given by
1
f XY ( x, y)  , 0  y  x  1
x
(a)Verify that this is a valid density for a two
dimensional random variable.
Exercise 5.1 (Q. 9)
1 x 1
1
(a )   dy dx   (1 / x )[ y ] y 0  1
x
0 0
x 0
(b) Find P [ X  0.5 and Y  0.25]
Sol : P[ X  0.5 and Y  0.25]

0.25 0.5
1
 
0 y
x
dx dy  0.4233

Exercise 5.1 (Q. 9)
(d ) Find P[ X  0.5 and Y  0.5]

1 1
1
   dx dy  0.15345
0.5 y
x

Examples: Continuous Joint Densities
 2 xy
 x  ; 0  x  1, 0  y  2
Ex.5.3. Let f  x, y    3
 0 ; e.w.
 1
(i) Find P  X   (ii) P Y  X 
 2

 1  2 xy 
1 2 1 2
Sol. (i) P  X      f  x, y  dydx     x   dydx
 2 12 0 12 0
3 
y 2 x 1
1
 2 xy 2
 2 2x 
1
2 3 x  5 2
  x y   dx    2 x   dx   x   
1 2
12
6  y 0 3  3 3  x 1 2 6
 2 xy 
1 x 1 x
(ii) P Y  X     f  x, y  dydx     x   dydx
0 0
0 0
3 
yx 1
1
 2 xy 
2
 3 x  1
 7x 
3
7 4
  x y   dx    x   dx    
0 
6  y 0 0
6  24  0 24

k  6  x  y  ; 0  x  2, 2  y  4
HW5.3. Let f  x, y   
 0 ; e.w.
(i) Find k
(ii) Find f X  x  , fY  y 
(iii) P Y  3
(iv) P  X  1  Y  3
(v) P  X  1 Y  3

2 4
Hint. (i)   f  x, y  dydx  1  k  1 8
0 2 y
4
(ii) f X  x    f  x, y  dy where 0  x  2; d
D
2
2 c
fY  y    f  x, y  dx where 2  y  4 x
0 a b
3
5
(iii) P Y  3   fY  y  dy  ,
2
8
1 3
3
(iv) P  X  1  Y  3    f  x, y  dydx 
0 2
8
P  X  1  Y  3
(v) P  X  1 Y  3 
3

P Y  3 5
kx 2 y; 0  x  1, 0  y  1
HW5.4. If f  x, y    is a pdf,
 0 ; e.w.
(i) Find k  Sol. k  6 
 1 1
(iii) P  X  Y  (iv) P  X  Y  
 2 2

k  3x  y  ; 1  x  3, 0  y  2
HW5.5. If f  x, y    is a pdf,
 0 ; e.w.
(i) Find k  Sol. k  1 28 
(iii) Are X and Y independent?

k ; 0  x  1, 0  y  x
HW5.6. If f  x, y    is a pdf,
 0 ; e.w.
(i) Find k  Sol. k  2 
(ii) Find f X  x  , fY  y  .

Problem Solving
HW.5.7. For the following pdfs, check whether X and Y are independent
 x 1  3 y 2 
 ; 0  x  2, 0  y  1
(i) f  x, y    4

 0 ; e.w.
2; 0  x  1, 0  y  x
(ii) f  x, y   
 0 ; e.w.
6 x 2 y; 0  x  1, 0  y  1
(iii) f  x, y   
 0 ; e.w.
1
  6  x  y  ; 0  x  2, 2  y  4
(iv) f  x, y    8
 0 ; e.w.
Problem Solving
Ex. 5.4: Suppose that the two dimensional

continuous random variable (X,Y) has
joint pdf given by:
 xy
x  , 0  x  1,0  y  2
2
f ( x , y)   3

0, elsewhere
Sol.
Check that
 
 
 
f ( x, y )
2 1
xy
   (x  2
) dxdy
0 0
3
 1 45 BITS Pilani, Pilani Campus
Sol.
Let B={X+Y1}, compute P(B).

1 1 x
xy 65
P(B)  1    ( x  ) dy dx 
2
0 0
3 72
P[B]= 1-P(B), where
y
2
B = (X+Y<1}, Hence
x
1
Cumulative distribution function
Def: Let (X,Y) be a two dimensional random
variable. The cdf F of the two dimensional
random variable (X,Y) is defined by:
F(x, y)  P[X  x and Y  y]
Remark: Function of two variables and has a

number of properties analogous to one
dimensional random variable.
CDF and PDF
If F( x , y ) is the cdf of a two dimensional
random variable with joint pdf, f ( x , y ),
 2
then F( x , y )  f ( x , y ),
x y
wherever F is differentiable.
(a) FX ( x )  F( x ,  ) and FY ( y )  F(  , y )
(b) X and Y are independent
 F( x , y )  FX ( x )FY ( y )
Multinomial Law
Ex 5.5: A die is rolled 10 times in succession. Find the
probability of the occurrence of six 4 times, five twice,
10
10! 1
and all other faces once each. Ans:  
4! 2! 1!1!1! 1!  6 
Ex 5.6:
An urn contains N  N 1  N 2   N m balls; N i : i th color
With replacement, n  i1  i2   im balls are picked.
Find P  X1  i1 , X 2  i2 , , X m  im 
i1 i2 im
n!  N1   N2   Nm 
      N 
i1 ! i 2 ! i m !  N   N   
Book Exercise 5.1 (Q. 12)
Ex. 12: Items coming off an assembly line are

classified as being either non-defective, defective but
salvageable, or defective but non-salvageable. The
probabilities of observing items in each of these
categories are 0.90, 0.08 and 0.02 respectively. The
probabilities do not change from trial to trial. Twenty
items are randomly selected and classified. Let X1
denote the number of non-defective items, X2 the
number of non-defective but salvageable and X3 the
number of non-defective but non-salvageable items
obtained.
Exercise 5.1 (Q. 12)
(a)Find P[X1=15,X2=3,X3=2].
(b)Find the general formula for joint density of
(X1, X2, X3).
20!
Soln : ( a)  0.90   0.08   0.02 
15 3 2
 15! 3! 2!

n!
( b ) f  x1 , x 2 , x 3   n1 n2 n3
p1 p2 p3
n1 ! n2 ! n3 !
(This is called multinomial distribution).
Expectation
Let  X , Y  be a two-dimensional RV with joint density f XY
and marginal densities f X , fY , respectively.
Expectation of H  X , Y  , denoted by E  H  X , Y   , is given as
E  H  X , Y      H  x, y  f XY ( x, y ), if  X , Y  is discrete RV.
all x all y
provided   | H ( x, y ) | f
all x all y
XY ( x, y ) exists.
 
E  H  X , Y      H  x, y  f XY ( x, y )dydx, if  X , Y  is continuous RV
 
 
provided   | H ( x, y ) | f XY ( x, y ) dy dx exists
 
Expectation: Properties
(a) E  c   c; c constant
(b) E  aX  bY  c   aE  X   bE Y   c
(c)E[a0  a1 X 1  ...  an X n ]
 a0  a1 E[ X 1 ]  ...  an E[ X n ]
If two random variables X and Y are independent,

E  XY   E  X  E Y  Verify!
Proof (for continuous case)
 
E  XY     xy f XY ( x, y ) dxdy
 
 
   xy
 
f X ( x) f Y ( y ) dxdy
  
 x

f X ( x)

yf Y ( y ) dy dx  x

f X ( x) E[Y ] dx

 E[Y ]  x f X ( x) dx  E[ X ] E[Y ]

Mean and Variance
Let  X , Y  be a two-dimensional RV with joint density f XY and marginal densities
f X , fY , respectively.
 X  E  X    xf XY ( x, y )   xf X ( x), if  X , Y  is discrete RV.
all x all y x
Y  E Y     yf XY ( x, y )   yfY ( y ), if  X , Y  is discrete RV.

all x all y y
  
X  E  X     xf XY ( x, y )dydx   xf X ( x)dx, if  X , Y  is continuous RV
  
  
Y  E Y     yf XY ( x, y )dydx   yf Y ( y )dy, if  X , Y  is continuous RV
  
Similarly,
 X2  E  X   x      x   X  f XY ( x, y )    x   X  f X ( x);  X , Y  discrete
2 2 2
all x all y x
  
 X2  E  X   x      x  X    x  X  f X ( x)dx;  X , Y  continuous
2 2 2
f XY ( x, y )dydx 
  

Examples
HW5.8. For a discrete random variable  X , Y  , P  X  i, Y  j   0, i, j  1, 0,1
1 1 1
except P  X  0, Y  1  , P  X  1, Y  1  , P  X  1, Y  1  .
3 3 3
Find E  X  , E Y  , E  X  3Y  ,Var  X  , Var Y  , Var  7 X  9 
HW5.9. Two balls are selected at random from a box that contains
3 blue balls, 2 red balls and 3 green balls. If X is the number of blue
balls and Y is the number of red balls selected, then
Find (i) f  x, y  (ii) f X  x  , fY  y  , (iii) E  X  , E  X 2  , E Y  , E Y 2 
, E  XY  (iv) Var  2 X  , Var  2Y  5 

Examples
k ; 0  y  x  1
HW5.10. f  x, y   
0 ; e.w.
Find (i) k (ii) f X  x  , fY  y  ,
(iii) E  X  , E  X 2
 , E Y  , E Y  , E  XY 
2
(iv) Var  X  , Var Y  , Var  XY  5 

Variance Properties
Ex.5.7. Show that Var  X  Y   Var  X   Var Y  ,in general
Sol. Var  X  Y   E  X  Y   X  Y  
 2
 
 E   X   X   Y  Y   
 2
 
 E  X   X   Y  Y   2  X   X Y  Y  
 2 2
 
 E  X   X   E Y  Y   2 E  X   X Y  Y  
2 2
 Var  X   Var Y  .
Here E  X   X Y  Y   is called the covariance.
Covariance
• It is a measure (absolute) of how much two variables
change together; e.g., height and weight, income-
expenditure, age-blood pressure, advertisement-sales,
demand-supply, fertilizer-yield.
• If two variables tend to show similar behaviour, then the
covariance is positive, otherwise negative.
• The sign of the covariance shows the tendency in the linear
relationship between the variables.
• The magnitude of covariance does not really produce a
fruitful meaning.

Covariance
Covariance provides a measure of joint variability.
If X , Y tend to show similar behaviour, then

the covariance is +ve, else it is -ve.
Let X , Y be RVs with means  X , Y , respectively.

Cov  X , Y   E  X  Y Y  Y  
 E  XY   E  X  E Y 
Covariance Properties
(i) If two random variables X and Y are independent, then Cov  X , Y   0
(ii) Cov  X , Y   Cov Y , X 
(iii) Cov  X , X   Var  X 
(iv) Cov  cX , Y   c Cov  X , Y  , c is a constant
(v) Cov  a  bX , Y   b Cov  X , Y 
(vi)Cov  a  bX , c  dY   bd Cov  X , Y 
  n 2
(vii)Var  a0   ai X i    ai Var[ X i ]  2 ai a j Cov  X i , X j 
n
 i 1  i 1 i j
 n
 n 2
(viii)Var  a0   ai X i    ai Var[ X i ],if X 1 ,..., X n are independent
 i 1  i 1
Remark on Covariance
If X and Y are independent, then Cov  X , Y   0,
but the converse is NOT true.
Let X be uniformly distributed in  1,1
and let Y  X . 2
Now Cov  X , Y   E  XY   E  X  E Y 
 EX 3
  E  X  E X 
2
0
Problem Solving
Ex.5.8. Let X , Y be independent RVs with means 1,  3, and variances
4, 5, respectively. Then, find E  5 X  3Y  9  ,Var  2 X  4Y  5  , Var  X  2 
Hint. E  5 X  3Y  9   5 E  X   3E Y   9  5  15  9  19
Var  2 X  4Y  5   22 Var  X   42 Var Y  , as X , Y are independent
Var  X  2   Var  X   4
kxy; 0  x  1, 0  y  x
HW5.11. Let f  x, y   
 0 ; e.w.
(i) Find k , E  X 2  , E Y 2  , E  XY  (ii) Var  X  , Var Y  (iii) Cov  X , Y 

Problem Solving
 x
1; x  1 and y 
HW5.12. f  x, y    2

 0 ; e.w.
Show that Cov  X , Y   0, but X , Y are not independent.
HW5.13. Let Cov  X , Y   K . Then, show that

Cov  aX  bY , cX  dY   0, if ac x2   ad  bc  K  bd y2  0.

Note
Sec 5.3 to be covered later. In Sec. 5.4, only

conditional densities will be discussed.

Conditional Density Functions
Let  X , Y  be a RV with joint density f XY and marginal densities f X , fY
(i) The conditional density for X given Y  y, denoted by f X y , is given by
f XY  x, y 
fX y  x  ; fY  y   0
fY  y 
(ii) The conditional density for Y given X  x, denoted by fY x , is given by
f XY  x, y 
fY x  y   ; fX  x  0
fX  x
Remarks:
(i) The conditional densities f X y and fY x are also density functions

Let  X , Y  : RV, joint pdf f XY , and marginal densities
f X , fY , respectively.
Notice that whenever X and Y are independent,

f X y  x   f X  x  and fY x  y   fY  y 

Examples: Conditional Density
 x 1  3 y 2 
    
Ex.5.9 f  x, y    4
; 0 x 2, 0 y 1

 0 ; e.w.
(i) Find f X  x  , fY  y  , f X y , fY x
(ii) Evaluate P 1 4  X  1 2 Y  1 3 .

1
x 1  3 y 2  x
Sol. (i) f X  x    dy  , 0  x  2;
0
4 2
x 1  3 y 2 
(If y0  [0,1], then
1 3y
2 2
fY  y    dx  ,0  y 1 fY (y 0 )  0
0
4 2
f XY  x, y0  x hence the conditional
f X y0  x    ,0  x  2
fY  y0  2 density f X |y0 (x) is
f XY  x0 , y  1 3y2 undefined for any x.)
fY x0  y    ,0  y 1
f X  x0  2
12 12
(ii) P 1 4  X  1 2 Y  1 3 
x 3
 f X Y 1 3  x  dx   dx 
14 14
2 64
2; 0  x  1,0  y  x
HW5.14. Let f  x, y    is a pdf,
 0 ; e.w.
(i) Find f X  x  , fY  y  , f X y , fY x and
(ii) evaluate P 1 4  X  3 4 Y  1 2  .

Hint. (i) f X  x   2 x,0  x  1; fY  y   2 1  y  ,0  y  1
f XY  x, y  1
fX y  x   , y  x  1;
fY  y  1 y
f XY  x, y  1
fY x  y    ,0  y  x
fX  x x
Note that f X y is a function of x, as y is a constant. Similar is the case for fY x .
1 1 1
(ii) When y  1 2, f X y  x   ,  x  1, that is, f X Y 1 2  x   2;  x  1
1  1 2  2 2
34 34
P 1 4  X  3 4 Y  1 2  
1
 f X Y 1 2  x  dx   2dx 
14 12
2

HW5.15. For the following pdfs, use conditional density
functions to check whether X and Y are independent
 2 xy
 x  , 0  x  1, 0  y  2
(i) f  x, y    3
 0 ; e.w.
  
sin x sin y; 0  x  , 0  y 
(ii) f  x, y    2 2
 0 ; e.w.
6 x 2 y; 0  x  1, 0  y  1
(iii) f  x, y   
 0 ; e.w.
Exercise 5.4 (Q. 54, Page: 188)
An electronic device is designed to switch

house lights on and off at random times
after it has been activated. Assume that
the device is designed in such a way that
it will be switched on and off exactly once
in a 1-hour period.

Let Y denote the time at which the lights

are turned on and X the time at which they
are turned off. Assume that the joint
density for (X,Y) is given by
f XY ( x, y)  8 xy , 0  y  x 1

(g) Find the conditional distribution of X

given Y
(h) Find the probability that the lights will
be switched off within 45 minutes of
the system being activated given that
they were switched on 10 minutes
after the system was activated.

Let X denote the number of “do loops” in a

Fortran program and Y the number of runs
needed to debug the program. Assume that
the joint density for (X,Y) is given as:
x/y 1 2 3 4
0 0.059 0.100 0.050 0.001

0.093 0.120 0.082 0.003
1
2 0.065 0.102 0.100 0.010
3 0.050 0.075 0.070 0.020
(d) Find the probability that a randomly

selected program requires at least two runs
to debug given that it contains exactly one
“do loop”
0.205
P[Y  2 and X 1]
Sol: P[Y  2 | X  1]  
P[ X 1]  0.688
0.298
(This can also be evaluated using fY x )
P[Y  2 | X  1]   fY | X 1 ( y ).
y2
Problem Solving
HW 5.16: The joint density of X and Y is given:
 1  xy
 ye , 0  x  , 0  y  2
f ( x, y )   2

0 , elsewhere
(1) Find E[X|Y=1]. (also denoted by X|y=1)
(this is the conditional mean of f X | 1)
(2) Find P[X>2|Y=1].
Solution
 
1  xy 1
fY ( y )  

f ( x, y )dx   ye dx  ; 0  y  2
0
2 2
 f ( x,1) x
 e ; 0 x
Hence f X |Y 1 ( x)   fY (1)
 0;
 elsewhere

E[ X | Y  1]   xf

X |Y 1 ( x)dx (Evaluate)

Solution

P[ X  2 | Y  1]   f X |Y 1 ( x)dx
2

x
  e dx (evaluate)
2

Course No: MTH F113: Probability and Statistics

Uploaded by

Copyright:

Available Formats

Course No: MTH F113: Probability and Statistics

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Course No: MTH F113: Probability and Statistics

Uploaded by

Copyright:

Available Formats

BITS Pilani

Course No: MTH F113

Probability and Statistics

3 BITS Pilani, Pilani Campus

4 BITS Pilani, Pilani Campus

5 BITS Pilani, Pilani Campus

Expectation and Moments

7 BITS Pilani, Pilani Campus

Single Random Variable (Univariate)

Two Dimensional Random variables (Bivariate)

9 BITS Pilani, Pilani Campus

Bivariate distribution occurs when we observe two

For example : Record the atmospheric temperature

• Let X and Y be discrete RV associated with a random experiment. Then the

Necessary and sufficient condition for a function to be a discrete joint density:

• For any region A in the xy plane, the probability of A is defined as

11 BITS Pilani, Pilani Campus

(ii) The marginal density for Y , denoted by fY  y  , is given by

Ex.5.1. For a discrete random variable  X , Y  , P  X  i, Y  j   0, i, j  1, 0,1

12 BITS Pilani, Pilani Campus

Sol (Ex.5.1). We represent pdf and marginal densities in a tabular form.

First we find out the events  x, y  such that x+y  1

14 BITS Pilani, Pilani Campus

• Let X and Y be continuous RV associated with a random experiment. Then

(i) The marginal density for X , denoted by f X  x  , is given by

(ii) The marginal density for Y , denoted by fY  y  , is given by

• The concept can be extended in an analogous

Let  X , Y  be a two-dimensional RV with joint density f XY

X and Y are independent if and only if

19 BITS Pilani, Pilani Campus

In general for n RVs,

Here f X1 ... X n ( x1 ,..., xn ) is the joint density

3. The joint density for (X,Y) is given by:

(b) Find the marginal densities for X and Y

23 BITS Pilani, Pilani Campus

4. The joint density for (X,Y) is given by

(c ) Are X and Y independent?

28 BITS Pilani, Pilani Campus

c) X and Y are not independen t

29 BITS Pilani, Pilani Campus

30 BITS Pilani, Pilani Campus

31 BITS Pilani, Pilani Campus

(9)An engineer is studying early morning

32 BITS Pilani, Pilani Campus

east – west direction. Time is measured in

(b) Find P [ X  0.5 and Y  0.25]

Sol : P[ X  0.5 and Y  0.25]

34 BITS Pilani, Pilani Campus

(d ) Find P[ X  0.5 and Y  0.5]

35 BITS Pilani, Pilani Campus

36 BITS Pilani, Pilani Campus

37 BITS Pilani, Pilani Campus

38 BITS Pilani, Pilani Campus

40 BITS Pilani, Pilani Campus

41 BITS Pilani, Pilani Campus

42 BITS Pilani, Pilani Campus

Ex. 5.4: Suppose that the two dimensional

Let B={X+Y1}, compute P(B).