PROBABILITY FUNCTION

AND
DISTRIBUTION FUNCTION

Presented
By
Pradip Panda
Asstt. Professor, Deptt. Of Statistics
Serampore College
 CONTENTS:

 Probability Function
Probability – mass function
Probability - density function
Joint probability – mass function
Joint probability – density function
Distribution function

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
 PROBABILITY FUNCTION

Probability – mass function (p.m.f.)

Probability – density function (p.d.f.)

With the help of these functions , the

probabilities of the random variables can be
found for different values or different sets of
values.

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
 PROBABILITY – MASS FUNCTION

For a discrete random variable X, there exists a function

f(x) such that f(x) = P(X = x), x being the general value
of X.

A function of this type satisfying the conditions:

i ) f ( x)  0, for all x,

ii)  f ( x)  1 ,
x
is called the probability – mass function (p.m.f.)
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
of X.College
Serampore
 e   x
f ( x)  P( X  x)  , x  0,1,2,......... ..;   0
x!
0 , otherwise

Is it a probability mass function?

Proof: Here, f ( x)  0 for all x.

 
e 
 x
and x 0
f ( x)  
x 0 x!
0  2  
 e  [
0!

1!

2!
 .......... .] e e =1
So, f(x)is a p.m.f.
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
Serampore College
 PROBABILITY - DENSITY FUNCTION

For a continuous random variable X, there may exist a

function f(x) such that, for a  b ,
b
a
f ( x) dx  P (a  X  b).

A function of this type satisfying the conditions:

i ) f ( x)  0, x

and ii)  f ( x) dx  1,


is called the probability – density function (p.d.f.) of

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
X.Serampore College
 Is the following a probability density function?
f ( x)  2 x , 0  x  1
 4  2x , 1  x  2
 0 , elsewhere
Proof: Here, f ( x)  0 , x and
 1 2

f ( x) dx   2 x dx   (4  2 x) dx
0 1

x2 1  x2 
 2. ]0   4 x  2.  
2
1
2  2
 1  [(8  4)  (4  1)]  2  1
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
So, f(x) is not a probability density function.
Serampore College
 JOINT PROBABILITY - MASS FUNCTION

For a pair of discrete random variables X and Y , a function f(x,y) , called

the joint probability – mass function of X and Y, such that

f(x,y) = P(X = x, Y = y) , x and y being the general values of X and Y,

and f(x,y) satisfies the conditions :

i ) f ( x, y )  0,  x, y

and ii)  f ( x, y )  1
x y

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
 JOINT PROBABILITY – DENSITY FUNCTION

For two continuous random variables X and Y , the joint

probability – density function, f(x,y), is such that :
d b

  f ( x, y) dx dy  P(a  X  b, c  y  d )
c a

and f(x,y) satisfies the conditions:

i ) f ( x, y )  0, x, y
 
and ii)   f ( x, y) dx dy  1


Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
 MARGINAL AND CONDITIONAL PROBABILITIES

DISCRETE RANDOM VARIABLE:

If f(x,y) be the joint probability – mass function of X and Y, then

the marginal distribution of X and Y are respectively given by :
g ( x)   f ( x, y ), and h( y)   f ( x, y)
y x
The conditional distribution or array distribution of X given Y= y,
is given by f ( x, y )
g ( x / y)  , provided h( y )  0
h( y )
The conditional distribution or array distribution of Y given
X= x, is given by
f ( x, y )
h( yProf.,
Mr. Pradip Panda,Asstt. )  of Statistics, , provided g ( x)  0
/ xDeptt.
Serampore College g ( x)
 Continuous random variable :

If f(x,y) be the joint probability – density function of X and Y,

then the marginal distribution of X and Y are respectively given
by :
g ( x)   f ( x, y ) dy and h( y)   f ( x, y) dx
y x
If f(x,y) = g(x).h(y) , for all x and y , then the variables X and Y
are said to be independent.
 
Cov( X , Y )  
  x    y    f ( x, y) dx dy
X Y
 
  xy f ( x, y ) dx dy   X Y
 
 
Where ,  X  Prof., Deptt. ( x) dx and Y   y h( y ) dy
x ofgStatistics,
Mr. Pradip Panda,Asstt.
Serampore College
 
 Is the following a joint probability – density
function? If yes, examine whether the variables are
independent or not.

f ( x, y )  x  1e , 0  x  1, y  0
2 y

3
 0, elsewhere.
Solution: Here, f ( x, y )  0, for all x, y and
   1
 
 
f ( x, y ) dx dy  
0
2
0 3  x  1e y
dx dy
1
 
  e dy.  x  1 dx   e 0 .  x   .1.  1
2  y 1 2 y  x 2
2 3
3 Mr. 0 Prof., Deptt. of Statistics,3
0 Pradip Panda,Asstt.
Serampore College
 2 0 3 2
 So, f(x,y) represents a joint p.d.f.

Marginal distribution of X is given by g ( x)   f ( x, y ) dy

 

2
x  1e dy  ( x  1) 0 e dy  ( x  1).1  ( x  1)
y 2 y 2 2
0 3 3 3 3

Marginal distribution of Y is given by h( y )   f ( x, y ) dx

   x  1e dx  e  ( x  1) dx  e . y
2 2 y 1 2 y 3

1
y
0 3 3 0 3 2
e
2
Thus, f ( x, y )  ( x  1)e  y  g ( x).h( y ) for all x and y.
3
So, the variables X and Y are independent.
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
Serampore College
If the joint p.d.f. is given as
 x y
f ( x, y )  e , x  0, y  0
 0 , elsewhere
Find the covariance between X and Y?

Solution: Marginal distribution of X is g ( x)   f ( x, y ) dy
 x
e x
 e y
dy  e , x  0
0
and g(x) = 0 , otherwise

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
 
Marginal distribution of Y is h( y )   f ( x, y ) dx

 y
e y
 e x
dx  e , y  0
0

and h(y) = 0 , otherwise

  
 E ( X )   x. g ( x).dx   x.e dx   e  x .x 21dx x
 0 0

 2  1
Similarly, E(Y) = 1
   
E ( XY )    xy f ( x, y ) dx dy    xy e  x  y dx dy
  0 0
 E ( X ).E (Y )  1
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
So, Cov(X,Y) = 0.
Serampore College
 DISTRIBUTION FUNCTION

It is useful in finding the quantiles and is denoted by F(x).

F ( x)  P ( X  x)
Properties: i ) 0  F ( x)  1
ii) F ()  0
iii) F ()  1
iv) F (a )  F (b) for a  b
and v) F ( x) is right continuous.
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
Serampore College
 For a continuous random var iable X with p.d . f .
f ( x),

Mean(  )  E ( X )   x. f ( x)dx

e
1
Median(  e ) is such that P ( X   e )   f ( x)dx  .

2

Mode(  0 ) is the value at which f ( x) is max imum



Variance( )  E[ X  E ( X )]   ( x   ) f ( x)dx
2 2 2

 
 x f ( x)dx   .
2 2
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
Serampore College

 
Mean deviation about ' a '  E ( X  a )   x  a . f ( x)dx


Moment of order r about ' a '   r' (a )  E ( X  a ) r

 r
  ( x   ) . f ( x)dx


The distributi on function F ( x) is defined as

x
F ( x)  P( X  x)   f ( y) dy

d
 Panda,Asstt.
f ( xMr.) Pradip [ F Prof.,
( x)]; F ( x) is a differentiable function.
Deptt. of Statistics,
Seramporedx College
For a discrete random var iable X with p.m. f . f ( x),

Mean(  )  E ( X )   x.P ( X  x)   x. f ( x).

x x

Median(  e ) is that value of the var iable which is

1
such that P ( X   e )   P ( X   e ).
2
Mode(  0 ) is the value of the var iable with the
highest probability.
Variance( )  E[ X  E ( X )]   ( x   ) . f ( x)
2 2 2

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

x
Serampore College
   ( x   ) . f ( x)   x . f ( x )   .
2 2 2

x x

Mean deviation about ' a '  E ( X  a )   x  a . f ( x).

x

Moment of order r about ' a'   (a)  E ( X  a) '

r
r

  ( x  a) . f ( x)
r

The distribution function F ( x) is defined as

F ( x)  P ( X  x)   f ( y ).
y x

Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,

Serampore College
Mr. Pradip Panda,Asstt. Prof., Deptt. of Statistics,
Serampore College