0

DOM
DIMENSİONAL
VARIABLES
SYLLABU
S:
Joint Distributions Marginal and
itional Distributions Functions
two dimensional random variables
— Regression Curve Correlation.

IFETCWH&S- II/MATHS/MATHIVADHANA/I YEAR/ M E (CSEj/I -SEM/MA7155/APPLIED PROBABILITY

AND 2
STATISTICS /UNIT—I I/PPT/VER1
 TWO - DIMENSIONAL
R.V.
Let be the sample space. Let
and Y(s) be two functions each
assigning a real number to each
outcome s c Then (X,Y) is a two
random variable.
imensional
The range of the imensional r.v.
two is (X,Y)
 RTEwVo i a e o wo
•ensiona
Two i en iona Discrete
es RV
• Two i en iona Continuous R
V
Two Dimensional Discrete R.V.:

the possible values of (X,Y) are fnite or

infnite,
countabthen (X,Y) in ca a two i en iona
r v When (X,Y) in a two- iecrete
e
ipossible values of (X,Y) may
en be repre
iona iecreteente as
r v the
(Xi j) '
1,2, . . n,. . , j - 1,2,. . , . .

Two Dimensional ontinuous R.V.:

If (X,Y) can accuse a values pecife re ion R
the
in aXY plane (X,Y) in ca e a in i en iona
continuous
two
IkL’CL9I%%IId‹1A’N5d‹1 UII*ALNAH d"LAl/l•1.L‹C5LlN

rv5L‹1d‹1AA%:/Al'LILLll†BABILI’"AHL5’AI5’|
Joint Distributions - Maroinal and Conditional
Distribution
Joint Probabili distribution of two random
X and Y
variables
Two Dimensional Discrete Random
Variables

Maroinal Probabili

Mass function

Conditional Distribution

PROBLEMS
Two Dimensional ontinuous Random
Variables

Maroinal Probabili Densi function .d.

onditional Distri ution

Cummulative Distribution Discrete & Continuous

Inde endent Random Variables Discrete &

Continuous

P BLEM
 Joint Probabilit distribution of two
random
variables and
We define the probability of joint event
which is a function of the numbers x and y by a
joint probability distribution function and denoted
by
 Two Dimensional Discrete Random Variables

Join P obabi i Mass Func on o and Y.

If (X,Y) in a two-dimensional discrete r v such

that P (x - X: , ' j) ' pij then pij iS Called the
p or simply the probability function of (X,Y)
provided the following conditions are ratiffed

The set of triples Xi j,pij i' 1 2. ., ,. ., j - 1,2,. ., n,. .,

in
called the joint probability distribution of (X,Y)
Maroinal Probabili Distribution:
The individual probability distribution of a random
variable is referred to as its marginal Probability
distribution.
Maroinal robabili mass function of X:
If the joint probability distribution of two random
variables X and Y is given, Then Marginal probability
mass function of X is given by

Then Marginal probability mass function of Y is given

by
 Conditional Probabili
Distribution

If (X,Y) are discrete random variables then,

Conditional robabilit distribution of X iven Y
-

Conditiona robabili distribution of iven X -

l t Y x:
Joint Probabili Densi Function:

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND

STAT I STICS /UNIT—II/PPT/VER1.2 1
1
 Conditional Probabili
Distribution

( ,Y) are continuous random variables then,

Conditional robabilit distribution of X iven Y
-

Conditional robabilit distribution of Y given X -

xc

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND

STAT I STICS /UNIT—II/PPT/VER1.2
Cumulative Distribution Function

The cdi of a discrete r v (X,Y) can be

written as

The cdi of a continuous r v (X,Y) can be

written as

F(X,Y) be the joint probability distribution

function then the joint probability density function
(x, ) in given by
 PROPERTIES OF DISTRIBUTION
FUNCTION
• F(- ,- ) = 0, F(- , y) = 0, F(x, - ) = 0
• F(- , ') = 1
• 0 x F(x, y=) 1
• F(x, y) is an increasing function of x and y.
(i.e.) if X « x2 and y « y2 then F(x , y ) x F(x2, y )
x F(x2, y2) and F(x , y ) x F(x , y2) F(x2, y2)
5. F(x, y) is right continuous in each of its
variates.
(i.e.)

6. P(a « X x b, Y x y) = F(b, y) - F(a, y)

7. P(X= x, c « Y x d) = F(x, d) - F(x, c)
8. If P(a « X x b, c « Y x d) = F(b, d) - F(a, d) + F(a, c) - F(b,
c)
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY
AND
STATI STICS /UNIT— II/PPT/VER1.2
nde enden Rando
Variable
Let X and Y be two random variables, discrete or
continuous, with the joint probability distribution (x,y)
and marginal distribution g(x) and h(y) respectively, the
random variable X and Y are said to be statistically
independent i and only if

for all ( , y) within their independent random variable

if(X,Y) are discrete then

if(X,Y) are continuous then

For the bivariate probability distribution of (X, F) given below, find PBX ñ l),
P{Y ñ 3), P(X £ 1, Y S 3), P(X 6 US 6 3), P(i’ £ 3?X ñ 1) and PBX + ¥ S 4).

1132 N32 2f32 3/32

6 /16 118 1/8 118
1/32 1?32

e(x « 1› = e(z= o› + s‹x -)

- g, s(x= o, r•;t + f,
p(x• 1, r=;›
 2
- 0+0+
32 2
32 32 16 16 8 8
1
32

= 0 + 1 1
lfi 0+
32 16 32
23 32
32 32 64
 0+0+
16
32 8
9ï 32
18
23 ï ó4
23
= žț r‹x= o, r=,ş › 9/žțeçx•
32 9i, Z r‹x= 2,
r-,j 7/8 28 r=i›
32 4 16
32
The joint protiability mass function of tY, F) is given by z, y) = @Zr + 3y),
z - 0, 1, 2:y= 1, 2, 3. Pind all they margiaai and cmiditional probability
distribu-

2 3
0 9£
1
2 13
*

7
2
 1
' ' ' "" 8
72
72
30
72
Total = 1

I3N7
24T2

Total = I
 C.P.D. o f f , givea F- 3

-
33 3
t31
2
t3
2
T33£ =l
”
33

CJ•.D. of F, 3ivm X-
0
r-i eJr
t8k ”

3 —9k

—I
IBM
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND
2 STAT ISTICS /UNIT— II/PPT/VER1.2 2
1
 CA.D. of P, givm k- (z+r
t )

1
2
1
2 3 @12 @2I
12
"
2
3 —13* — #z 2 ' 712
*
13
Total - I
” 3 " 12
Total= i

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND

STAT ISTICS /UNIT— II/PPT/VER1.2
2
 PROBLEMS ON TWO DIMENSIONAL CONTINUOUS RANDOM
VARIAB PROBLEM 1:

Here the rectangle defined by 0 A z £ 2, 0 A y A I is the raage space A. A ,

fig ...

I9

IFETCMH&S II/MATHS/MATHIVADHANA/I YEAR/ M E (CSEj/I SEM/MA7155/APPLIED PROBABILITY

AND 2
STATISTICS /UNIT—I I/PPT/VER1 2
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY
AND 2
STATI STICS /UNIT— II/PPT/VER1.2 4
 i›r‹ 2
1T4 6
—

/ ' ( I » I) 19Z 24
19
)
dxdy

48
0

IFET CE-/H&S II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I SEM/MA7155/APPLI ED PROBABILITY AND

STAT ISTICS /UNIT— II/PPT/VER1.2 2
5
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY
AND 26
STATI STICS /UNIT— II/PPT/VER1.2
PROBLEM
2:

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY

AND 2
STATI STICS /UNIT— II/PPT/VER1.2 7
 !kiottum›g*«p«iéttua›ee«iibisdneiu§eOA8;âx›wiiatlnfig¡im),
&timl 8y 0<z<1 xy <x

=— b 0 «x«2
4
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI
ED PROBABILITY AND 28
STAT ISTICS /UNIT— II/PPT/VER1.2
 3y
(d =(
)

.J’ l

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY

AND 2
STAT ISTICS /UNIT— II/PPT/VER1.2 9
FUNCTIONS OF TWO DIMENSIONAL
R.V.:

W’= fi(X, I’} are two ott›az RYa, tbea tIzsj‹ñst pdf of(Z, K'} is
givsa by
_ â(a. y)

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND STATI

STICS / 3
UNIT—II/PPT/VER1.2 0
Problems on two dimensional random variables :

() W A M x RVwflAomaaxesnandv .NmMAapdfo
[C) IXX 4B#m?buboow?tR pfzed Tw
sn

s‹d cb is tbnpd€of a b›goor zeal

dTstzitn›ttoo.

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND STATI

STICS / 3
UNIT—II/PPT/VER1.2
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY
AND
STATI STICS /UNIT— II/PPT/VER1.2
 Stac and I are independentnormal Res (z, i
k < z, / =
< •. The joint pdf of (A l¥) is gi¥en
by
2 8
The pdf otZ is the marginal pdf, otitained by integrating /„ ‹c, w) with
respe
to n- over the range of w.

16
*

which is N (0, 2 13 )
•
lfX and F each follow an exponential distribution with parameter 1 and arc inde-
pendent, find the pdf of U - X ¥.
/,f r) = c °. .r > 0. and /,(y) = e ", v >0
Since X and r are independent.

Consider the auxiliary RV V = 7 along widi

U X - Y.

= 1

The joint pdf of I t/, V I is given by

The range space of (£/. Y ) is found our from ibe of the range space of
(A. D under the transformations z - u + c and y = v.

IkL’CL/I II/l1. ’II5/l1. ’III.4 UL.J//l1.L.‹CRL:/I %Ll1/l1. /’::6J--LILL l-/†B.

US LI h4.l LILI’°.UL j<
%’ ’|%’|C%/Lfl|’-
 p. ,
’

Therefore. the range space of (9. is given by r > — x, when « < 0 and r
when u > 0. >0,
Now the pdf of U is given by

f Iu1 -- when u <

e‘ 0
“ * ’“"
an dr, = r ’"“ ' i ” dr, wkcn u >
d 0

f u1 - e , when « < 0
2
Re ression:
It is a mathematical measure of the average
relationship between two or more variables in terms of
the original limits of the data.

The line of regression of y on

x is

The line of regression of x on

y is

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY 36

AND
Note: Both the lines of regression pass
through
Re ression
Coefficients:
Regression Coefficients of y on x
is

Regression Coefficients of x on y

is An ie between the re ression

lines: The slopes of the

regression lines are
If e is the angle between the regression
lines, then
Problems on Regression:

a). We
have 2

I 0—
x x
2

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY

AND 38
STATI STICS /UNIT— II/PPT/VER1.2
 b). Repression curve of Y on A is

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND STAT ISTICS /

UNIT— II/PPT/VER1.2
 2(x-+-1)

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND STAT ISTICS /

UNIT— II/PPT/VER1.2 40
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND STATI
STICS / 4
UNIT—II/PPT/VER1.2 1
 3 24
6 0
3
6 2M
1 14 324 IP6
8

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY

AND 42
STAT ISTICS /UNIT— II/PPT/VER1.2
 -(1A.17)
w (2035)
(6.44)
w (13.63)
* —0.93 x ’” —
b m0.43 13.6
”
z• 3

x=
.43yi23.S2

IFETCWH&S- II/MATHS/MATHIVADHANA/IYEAR/ M E (CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND

4
STATISTICS /
UNIT—I I/PPT/VER1.2 3
 CORRELATIO
Correlation: N
the two variables vary such that the change
in one variable affect the change in the other
variable The variable are said to be correlated

Eg: There exist more relationship between height

and weight of the person

Correlation Coefficient:
it in the parameter which measures the strength
and
direction of a relationship between two variables
To define the ovariance, we need to descri e
the expected value of a function of two random
variables

Ex e ed Va ues o a Two ens ona RV

If (X,Y) in a two- i en iona r v with joint probability

mans
function pij then

If (X,Y) in a two- en iona continuous r v with joint

p
i
x, ), then
Covariance :
It is defined as the useful measure to study the
relationship
between the variables.
=E E(x)

Note:
If X and Y are independent , then Cov[x,y] = 0
If X and Y are independent ,then E[xy] = E[x] E[y]

Variance:

If X and Y are random variables ,then

 Results
•: Cov [ax,by] = ab cov [x,y]
• Var [X+Y] = Var [x] + Var [y] + 2 Cov [X,Y]
• Cov [X+a,Y+b] = Cov [X,Y]
• Var [aX+b] = a2 Var [x]

T es o Corre ation
1 Positive and Negative
2) Simple, Partial and
Multiple 3 Linear and Non
Linear

• Positive Correlation: If the two variables deviate in the

same direction (ie) if the increase (decrease) of one
variable results in the corresponding increase
(decrease) of the other variable

Eg: Correlation between rainfall and production are

positive
Ne ative Correlat
If the two variables deviate in the opposite direction. (i.e.)
on
if the increase (decrease) of one variable results in the
corresponding decrease (increase) of the other variable.
Eg: Correlation between Price of a commodity and its
demand are negative

2 im ie Pa ial and Mult ie Correlat on

If only two variables are considered for correlation Analysis
it is called si ie correlat on
Eg: Height and Weight of students in a
class.

If two or more variables are considered for

correlation Analysis ,it is called Multi ie Correlat
on
Eg: In the study of relationship between the yield of
rice ,both the amount of rainfall and usage of
fertilizers are considered.
3 Linea and Non L nea Co e a on
the amount of chan e in one variable ten to
bear the constant ratio to the amount of chan e
in the other variable, then the correlation in said to
be linear
the amount of chan e in one variable oe not
bear the constant ratio to the amount of chan e in
the other variable, then the correlation in said to
be Non-linear
or Curvilinear
: rain in more then the pro uction of
nota ari bewou
rice oub
nece e
Methods of studvin correlation:
• Scatter Diagram Method
• raphic Method
• Concurrent Deviation Method
• Method Of Least uares
• Karl Pearsons Co-efficient Of Correlation
• Rank Method
POSITIVE EAR
LI CORRELATION

(a) ng (c)
Positive positiv Perfec
e t
positiv
Scaber e
Plots
IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY AND STATI
STICS / 5
1
NEGAT LINEAR LATION

(d) (e) (f}

Negative Strong Perfect
negativ negativ
e e

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY

AND 52
STATI STICS /UNIT— II/PPT/VER1.2
 NON LINEAR CO LATION

(g) No (h) Nonlinear Correlation

Correlation

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND STAT ISTICS /

UNIT— II/PPT/VER1.2 S
3
 CO I.ATION

• Measures the relative strength of the

linear
relationship between two variables
• Unit-been
• Ranger between —1 and 1
• The closer to —1, the stronger the
negative linear relationship
• The closer to 1, the stronger the positive
linear relationship
• The closer to 0, the weaker and
positive linear relationship
SCATTER PLOTS OF DATA WITH
LATION
COEFFICIENTS

IFET CE-/ H&S II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I SEM/MA7155/APPLI ED PROBABILITY

AND
55
Results:
• The value of a correlation coefficient ranges between -
1 and

• Two Independent Random variables are said

to be uncorrelated.

• When X and Y are independent ,then cov

=0

• -1 s r s 1

• If r » 0 we have a Positive correlation

• If r 0 we have a Negative correlation

• If r = 0 we have No correlation
KARL PEARS0N'S CO-EFFICIENT
OF CORRELATION:

measures the strength of the linear

relationshi between the variables.
Correlation Coefficient between two random
variables
and is usual denoted
and
defined as
If (xc ,yr) i =1,2,...,n is the bivariate distribution, then

Where n is the no of items in the

given data.
IFETCWH&S- II/MATHS/MATHIVADHANA/I YEAR/ M E (CSEj/I -SEM/MA7155/APPLIED PROBABILITY AND
STATISTICS / S
UNIT—I I/PPT/VER1 2
IFETCE/H&S- II/ktATHS/ktATHlVAbHAH IYEAR/ M.E.(CSE}/I-SEk1/MA7155/APPLIEb PROBABILITY
AHb 5
 IT
=68x69=4692

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLI ED PROBABILITY AND STAT ISTICS /

UNIT— II/PPT/VER1.2 60
RANK CORRELATION
:
Let us suppose that a group of n individuals
are arranged in order of merit in possession of
two characteristic A and B There
ranks with, in general, be different

pear an’ Rank Correlation Coe cient in given

by
Problems based on Rank Correlation
:
Find the rank correlation coe cient
between the following data

Rank in 2 3 4 5 6 7
X
Rank in 4 3 2 6 5 7
Y
Solution:
X Y
1 4 -3 9
2 3 -1 1
3 1 2 4
4 2 2 4
5 6 -1 1
6 5 1 1
7 7 0 0

Rank correlation coefficient is given

by

IFETCE-/H&S- II/MATHS/MATHIVADHANA/IYEAR/ M E.(CSE}/I -SEM/MA7155/APPLIED PROBABILITY

AND 6
STATI STICS /UNIT— II/PPT/VER1.2 3
Re eated
Ranks
there in more than one item with the have
value in the Series, then the usual etho cannot
be use , to common ranks are iven to repeate
ranks are iven the to owin Correction
in ma e in the correction formula

in the correlation formula, we a a factor to

where in the number of items item in
an correction factor in to be a repeate
This
for each
e value repeate
Problems on Re eated ranks
Find the rank correlation coefficient from the
following
X data
68 64 75 50 64 80 75 40 55 64
Y 62 58 68 45 81 60 68 48 50 70

X Y Rank X Rank Y

68 62 4 5 -1 1
64 58 6 7 -1 1
75 68 2.5 35 -1 1
50 45 9 10 -1 1
64 81 6 1 5 25
80 60 1 6 -5 25
75 68 2.5 35 -1 1
40 48 10 9 1 1
55 50 8 8 0 0
64 70 6 2 4 16
in X Series
to common rank 25 in given for each
75 to common rank 6 in given for
each 64 to common rank 3 5 in
given
- 10for each 68

Repeated value Positions Average

ranks

75 2&3 (2+3)/2=2.5

64 56&7 (5+6+7)/3=6

68 3&4 (3+4)/2=3.5