CE 207

Applied Mathematics for Engineers

Course Teacher: Dr. Md. Hadiuzzaman
Room No. 544, Civil Building
(Lecture 3)
 Joint Probability Mass Function
p(xi , y j ) = PX = xi , Y = y j 
• X an Y are both discrete random variables whose possible
values are, respectively, x1, x2, x3, ...... etc. and y1, y2, y3,
......etc.
• Individual probability mass functions of X and Y are easily
obtained from Joint probability mass function

P( X = xi ) =  PX = xi , Y = y j  =  p(xi , y j )
j j

P(Y = y j ) =  PX = xi , Y = y j  =  p(xi , y j )

i i
 Problem 1
Q. 3 batteries are chosen from a group of 3 new, 4 used but working, 5
defective batteries. Determine probabilities of different combinations.

Soln:
Let X and Y denote respectively the number of new batteries and used but
working batteries that are chosen. Then probable values of X are 0, 1, 2, 3
and probable values of Y are 0, 1, 2, 3
x1,x2,x3,x4 y1,y2,y3,y4
p(xi,yj) denotes joint probability that xi no. new batteries and yj no. used
but working batteries will be chosen.

p(0,0) = 5C3 12
C3 = 10 220
( )( C )
p(0,1) = 5C2 4
1
12
C3 = 40 220
 Problem 1 (Contd….)

p ( 0, 2 ) = ( 5C1 )( 4C2 ) 12
C3 = 30 220
p ( 0,3) = 4C3 12
C3 = 4 220
p (1, 0 ) = ( 3C1 )( 5C2 ) C3 = 30 220
12

p (1,1) = ( 3C1 )( 4C1 )( 5C1 ) C3 = 60 220

12

p (1, 2 ) = ( 3C1 )( 4C2 ) 12

C3 = 18 220
p (1,3) = 0
 Problem 1 (Contd….)

p ( 2, 0 ) = ( 3C2 )( 5C1 ) 12
C3 = 15 220
p ( 2,1) = ( 3C2 )( 4C1 ) C3 = 12 220
12

p ( 3, 0 ) = ( 3C3 ) C3 = 1 220
12

Also p ( 2, 2 ) = p ( 2,3) = p ( 3,1) = p ( 3, 2 ) = p ( 3,3 ) = 0

 Problem 1 (Contd….)
Probability that no new batteries will be chosen is:
P ( X = 0 ) =  P  X = 0,Y = y j  =  p ( 0, y j )
j j

= p ( 0 ,0 ) + p ( 0 ,1) + p ( 0 , 2 ) + p ( 0 ,3 )
= (10 + 40 + 30 + 4 ) 220 = 84 220
Similarly, probability that one/two/three new batteries will be chosen is:
P ( X = 1) = p (1, 0 ) + p (1,1) + p (1, 2 ) + p (1,3) = 108 220
P ( X = 2 ) = p ( 2, 0 ) + p ( 2,1) + p ( 2, 2 ) + p ( 2,3) = 27 220
P ( X = 3) = p ( 3, 0 ) + p ( 3,1) + p ( 3, 2 ) + p ( 3,3) = 1 220
Note, sum of all these probabilities i.e., P(X=0,1,2,3)=1
 Problem 2
Q. 15% of families have no children, 20% have 1 children, 35% have 2
children, 30% have 3 children. Girl or boy is equally likely. What will be the
joint probability mass function of a family if it is chosen at random from
this community?

Soln:
Let B and G denote respectively the number of boys and girls respectively
in a family randomly chosen. Then probable values of B are 0, 1, 2, 3 and
probable values of G are 0, 1, 2, 3
b1,b2,b3,b4 g1,g2,g3,g4
p(bi,gj) denotes joint probability that bi no. boys and gj no. girls.

p(0,0 ) = P{B = 0, G = 0} = 0.15

p(0,1) = P{B = 0, G = 1} = P1 child P1 girl | 1 child  = 0.2(0.5) = 0.1
 Problem 2 (contd….)
p(0,2) = P{B = 0, G = 2} = P2 childrenP2 girl | 2 children
= 0.35(0.5)(0.5) = 0.0875
p(0,3) = P{B = 0, G = 3} = P3 childrenP3 girl | 3 children
= 0.3(0.5)(0.5)(0.5) = 0.0375

Probability that family has no boy i.e.,

P(B=0) = p(0,0)+p(0,1)+p(0,2)+p(0,3)=0.375

Similarly, probability that family has one girl i.e.,

P(G=1) = p(0,1)+p(1,1)+p(2,1)+p(3,1)=0.3875
where, p(3,1)=0
Joint Probability Mass Function

j 0 1 2 3 Row sum
= P(B=i)
i

0 .15 .10 .0875 .0375 .3750

1 .10 .175 .1125 0 .3875

2 .0875 .1125 0 0 .2000

3 .0375 0 0 0 .0375

Column .3750 .3875 .2000 .0375

sum=
P(G=j)
 Problem 2 (contd….)
PB = 0, G = 1 0.1 8
P{B = 0 | G = 1} = = =
PG = 1 0.3875 31
PB = 1, G = 1 0.175 14
P{B = 1 | G = 1} = = =
PG = 1 0.3875 31
PB = 2, G = 1 0.1125 9
P{B = 2 | G = 1} = = =
PG = 1 0.3875 31
PB = 3, G = 1 0
P{B = 3 | G = 1} = = =0
PG = 1 0.3875
Hence, conditional probability that family chosen has at least one boy, given
that there is one girl is = (14+9)/31=23/31
 Continuous Random Variable
• A continuous random variable X has an infinite (or nearly infinite) set of
possible values in a given interval
• There exists a nonnegative function f(x) defined for all real values of x
having the property that for any set B of real numbers

PX  B =  f (x )dx
B

• f(x) is known as probability density function of X

• In words, the probability that X will be in B may be obtained by integrating
the probability density function over the set B.
• All probability statements about X can be answered in terms of f(x)
• For RVs that map to the integers or the real numbers, the cumulative
density function (cdf) is a useful alternative representation
 Probability Density Function (PDF)

f (x)


PX  (− ,  ) =  f ( x ) dx = 1
−

• For RVs that map to the integers or the real numbers, the cumulative
density function (cdf) is a useful alternative representation
 Probability Density Function (contd.)
• Considering B={a,b}, we can write

Pa  X  b =  f (x )dx
b

a
• Putting a=b in above equation,

PX = a =  f (x )dx = 0
a

a
Hence, probability that a continuous random variable ‘X’ will
assume a particular value ‘a’ is zero!
 Probability Density Function (contd.)
• However, probability that a continuous random variable ‘X’ will be near a
particular value ‘a’ is not zero!

• Putting b=a+dx in above equation, where dx is small

Pa  X  b =  f (x )dx
b

a
a + dx
Pa  X  a + dx =  f (x ) dx  f (a ) dx
a
 Problem 1
f(x) = C(4x-2x2) for 0<x<2 

= 0 for other values of x

−
 f ( x)dx = 1
Determine C and P{X>1}


−


0
2
( )
f ( x )dx = C  4 x − 2 x 2 dx = 1 by definition of f(x)
x=2
 2 2x  3
i.e., C 2x −  =1
 3  x =0
3
i.e., C =
8
Now, PX  1 = 


1
(
3 2
8 1
)
f ( x )dx =  4 x − 2 x dx =
2 1
2
Cumulative Distribution Function (CDF)

F (a ) = PX  (− , a ) = P{ X  a} =  f (x ) dx
a

−

d
F (a ) = f (a )
Fundamental
theorem of
da calculus
 Cumulative Distribution Function (contd.)
Notes:
• F()=1 and F(-)=0
• F(x) is a continuous function even if f(x) is
discontinuous.
• As x increases F(x) either increases or
remains constant.
• P{c<X<d} = F(d)-F(c)
• P{X>a} = 1-F(a)
 Joint Probability Density Function
• Random variables X and Y are jointly continuous if there exists
a function f(x,y) defined for all real x and y having the
following property:

PX  A, Y  B =   f (x, y ) dx dy
B A

• If X and Y are jointly continuous, they are individually

continuous, and their individual probability density functions
f(x) and f(y) can be obtained as:
 

f (x ) = f (y) = 

−
f ( x, y )dy
−
f ( x, y )dx
 Problem 1
f(x,y) = 2e-xe-2y for 0<x<, 0<y<
= 0 for other values of x and y
Determine P{X>1, Y<1}, P{X<Y}

PX  1, Y  1 = 
1

0 1

− x −2 y
2e e
1
dxdy =  2e − 2 y − e − x
0
( 
1 )dy
 −2 y
 -1
1
= e -1  2e − 2 y dy = 2e -1 
e 1
(
 = e 1 − e − 2 )
 − 2y
0
0


PX  Y  =
y
 2e e dxdy =  
− x −2 y
2e − x e − 2 y dxdy
0 0
( x , y ): x  y

0

=  2e −2 y
(1 − e ) −y 2 1
dy = 1 − =
3 3
 Joint Probability Density Function (contd.)
• Random variables X and Y are independent if :

PX  A, Y  B = PX  APY  B

• In other words, X and Y are independent, if for all A and B, the
events {XA} and {YB} are independent.
• For the jointly continuous case, the condition is equivalent to

f ( x, y ) = f ( x ) f ( y )
• X and Y are independent if knowing the value of one does not
change the distribution of the other.
 Problem 2
Q. X and Y are independent random variables with common
density function: f(x) = e-x for x>0
= 0 for x0
Determine density function of X/Y.
FX / Y (a ) = PX Y  a =  f (x, y )dxdy =  f (x ) f ( y )dxdy
x / ya x / ya

 (1-e )e
 ay 
= 
−x − y
e e dxdy = -ay −y
dy
0 0 0
y =
− ( a +1) y
 −y e  1
= - e −  =1 −
 a + 1  y =0 a +1

f X / Y (a ) = (FX / Y (a )) = 1 2
d
da (a + 1)