QM PDF
QM PDF
QM PDF
VARIABLES
PART-I
Prepared by
Dr. A.R. VIJAYALAKSHMI
1
CORRELATION AND REGRESSION
Correlation:
1
X (X i X )2 (standard deviation of X)
n i
1
Y (Y i Y ) 2
n i (standard deviation of Y)
3
Karl Pearson correlation co-efficient
1
n i
( Xi X )(Yi Y )
xy
r( X , Y )
2 2
1 2 1 x y
( X i X ) (Yi Y )2
n i n i
Note
The following formula may also be used to compute correlation co-efficient
between the two variables X and Y.
N XY X Y
(i )r ( x , y ) 2 2
N X ( X ) N Y (Y ) 2 2
Nd d d d
(ii)r(x, y) x y x y
Where
2 2
Nd (d ) Nd (d )
x x
2
y y
2
1 2 2 1 2 2 Cov ( u , v )
u
n
u u , v
n
v v r ( X , Y ) r (u , v )
uv
Solution : X
X 544
68 Y
Y 552
69
n 8 n 8
X Y x X X y Y Y x2 y2 xy
6
r( X ,Y )
xy
24
0.603
2 2
x y 36 54
25(520) (125)(100)
0.667
2 2
25(650) (125) 25(436) (100)
7
Spearman’s formula for the rank correlation
When the numerical values of X&Y are not available, but the position
of the values are arranged in order of merit (ranks), then also we can
find the correlation coefficient between the characteristics of group of
individuals using the formula
n
2
6 d i
i 1 di xi yi , xi ' s , yi ' s Indicates ranks
rs 1 - 2
n(n - 1)
8
3.The table lists 7 schools and provides data about the percentage of pupils
who have free school meals and their GCSE results. Calculate
Spearman’s rank correlation coefficient and explain what this means
in the context of this question.
9
Solution: First rank the data:
10
Using the formula 6 d 2 6x96 576
rs 1 1 1 0.714
n(n2 1) 2
7(7 1) 336
This value represents negative rank correlation, so the schools
with the highest proportion of pupils receiving free school meals
tend to have the least successful GCSE results (and vice versa).
m ( m 2 1)
to be added to d 2
12
where m is the no. of item is repeated. This c.f should be added for each
repeated value.
11
4.Obtain the rank correlation coeff. For the
following data X:68 64 65 50 64 80 75 40 55 64
Y:62 58 68 45 81 60 68 48 50 70
Solution:
X Y rank in rank in d xi y i d2
xi yi
68 62 4 5 -1 1
64 58 6 7 -1 1
75 68 2.5 3.5 -1 1
50 45 9 10 -1 1
64 81 6 1 5 25
80 60 1 6 -5 25
75 68 2.5 3.5 -1 1
40 48 10 9 1 1
55 50 8 8 0 0
64 70 6 2 4 16
12
72
In X series 75 is repeated twice which are in the position 2 & 3 in ranks
567
common rank is 3
6 , to be given for each 64
13
Correction factor
2 1 1
6( d 2 )
2 2 6(72 5) 450
r 1 2
1 1 0.5454
10(10 1) 10 99 990
14
Regression Lines
For the pair of values of (X, Y), where X is an independent variable and
Y is the dependent variable the line of regression of Y on X is given by
. Y Y b yx ( X X )
where byx is the regression co-efficient of Y on X and given by
y
b yx r
x
Where x X X, y Y Y
15
Similarly when Y is treated as an independent variable and X as
dependent variable, the line of regression of X on Y is given by
X X b xy (Y Y ) where bxy
bxy r
x
xy
2 Where x X X, y Y Y
y y
16
Problems
5.Out of 2 reg lines given by x 2 y 5 0 &
2 x 3 y 8 0 . Which one is the reg.of x on y ?.
Solution:
Suppose x 2 y 5 0 is the equn. of the reg.
line of x on y & 2 x 3 y 8 0 is the equn.
of the reg. line of y on x ,
then the 2equns can be written as x 2 y 5
2 8 2
& y x
3 3 Hence b 3 & b 2
yx xy
4
Now r 1
2
3
17
This is impossible. Hence our assumption is wrong
2 x 3 y 8 0 is the equn. Of reg.
line of x on y
6. If and the angle bet. 2 reg. lines
x y
1
is tan 3. Obtain the values the correlatio n coefficien t.
Solution: The acute angle bet. 2 reg. lines is
1 r 2
1 r 2
2
tan
1
( ) x y
tan 1 ( )
2 2
2
r x
2
y
r
1 r
2
tan 1
tan 3 (given)
1
2 r
1 r 2
3 r 2 6r 1 0
2r
6 36 4 6 40
r 3 10
2 2
3 3 . 16 0 . 16 or 6 . 16 0 . 16
byx
xy 93
0.664
2
x 140
20
(i) Regression equation of Y on X is
Y Y b yx ( X X )
Y 38 0.664(X 32)
Y 59 .25 0 .664 X
ii)To estimate the marks in statistics (Y) for a given marks in the
Economics (X), put X = 30, in the above equation we get,
Y 59 . 25 0 . 664 ( 30 )
59 . 25 19 . 92 39 . 33 or 39
21
, 8. If y 2 x 3 & y 5 x 7 are the two regression
lines, find the mean value s of x and y
Find the correlatio n co - efficient between x & y .
Find an estimate of x when y1
Solution:
Given y 2 x 3 (1) y 5 x 7 ( 2 )
Since both the lines of regression passes through t he mean value s
x and y the point, x , y must satisfy the two given regression lines.
y 2 x 3 (3) & y 5 x 7 ( 4 )
Subtracting the equations (3) and (4), we have
10 10 29
3 x 10 x y 2 3
3 3 3
23
Properties of regression coefficients
24
25
Sample space and events
Random experiment :
An experiment in which all possible outcomes are known in advance
and the exact outcome in any specific trial of the experiment is
unpredictable is known as random experiment .
26
Example 1 The experiment consists of tossing two coins, all the possible
Outcomes of this experiment form the following set:
S (H, H), (H, T), (T, H), (T, T)
27
The subset of the set S is called event. The empty set
of S is called impossible event, and S is called certain event or sure event
Ex. In drawing a card from a pack of cards, the no. of cases favourable
to the event of getting an ace is 4 , for drawing a spade is 13.
Ex. In tossing a coin, the event of getting a head in the first toss is
independent of getting a head in the second, third & subsequent throw.
Complementary events:
Let E be an event of a random experiment
and S be its sample space. The set containing
all the other outcomes which are not in E but in the
sample space is called the compliment ary event of E.
It is denoted by E . Thus, E S E . Note that E and E
are mutually exclusive events
29
In throwing a die, let E 2, 4, 6} be an event of getting
{
a multiple of 2. Then the complement ary of the event
E is given by E 1,3,5
Probability(Mathematical or classical)
If random experiment or a trail results in ‘n’ exhaustive, mutually exclusive
and equally likely outcomes(or cases),out of which m are favourable to the
occurrence of an event E, then the probability ‘p’ of occurrence
(or happening) of E, usually denoted by P(E), is given by
30
No .of favourable cases m
p P(E)
Total no .of exhaustive cases n
Note:
31
(iv) The probabilit y of an impossible event is 0.
That is P ( ) 0
(vi) P (A) P (A ) 1
32
Problems
9.How many elements will be there in S
5
n( E ) 5 & P ( E )
36
E ( 4 ,5 ), ( 5 , 4 ), ( 3, 6 ), ( 6 ,3 ), ( 5 , 6 ), ( 6 ,5 )
6
n(E ) 6 & P (E )
36
34
iv) Probability of both odd digits
9
n( E ) 9 & P ( E )
36
(1,1),(1,3),(1,5),(2,2),(2,4),(2,6),(3,1),(3,3),(3,5),
E
(4 ,2), (4,4), (4,6), (5,1), (5,3), (5,5), (6,2), (6,4), (6,6)
18
n ( E ) 18 & P ( E )
36
vi) Multiples of 3 as sum
35
(1,2), ( 2,3), (1,5), (5,1), ( 2,4), ( 4,2), ( 4,5), (5,4), (3,3),
E
( 3, 6 ), ( 6,3), ( 6, 6 )
12
n( E ) 12 & P ( E )
36
vii) a doublet
1
n( E ) 6 & P ( E )
6
11. What is the probability that an ordinary year has 53 Sundays ?
36
12.A person is known to hit the target in 3 out of 4 shots.where as an
other person is known to hit the target in 2 out of 3 shots .Find the
Pro. of the target being hit at all when the both person try.
3 1
P ( A) , P ( A )
4 4
1 1 1
P ( Not hitting the target) P ( A B ) P ( A ) P ( B )
4 3 12
37
Problems on independent events:
1 1
Solution: Given P (A)
6
, P (B ) ,
4
1 5 1 3
P( A) 1 , P(B ) 1
6 6 4 4
1
P ( both selected ) P ( A B ) P ( A ) P ( B )
24
st
4C 1 4
P ( 1 ball is white)
10 C 1 10
nd 6
P( 2 ball is black )
10
st nd 4 6 6
P ( 1 ball is white & 2 ball is black )
10 10 25
41
Dependent events:
Two events A&B are said to be dependent events when B
can occur only when A known to have occurred
(or vice versa).Probability attached to such events
are called conditional probability.
.
Conditional Probability :
The prob. of an event A given that B has already occurred
is called the conditiona l prob. of A given B is
P ( AB )
P ( A / B) , P ( B ) 0 Also conditiona l prob. of
P (B)
P ( AB )
B given A is P ( B / A) , P ( A) 0
P ( A)
42
Without replacement (dependent event)
1.A bag contains 25 tickets, a ticket is drawn and without
replacement another ticket is drawn. Find the prob. that
both tickets will show even no.s.
Solution: S 1, 2 , 3 , 4 , 5 ,... 25 , n ( s ) 25
E 2 , 4 , 6 ,8 ,10 ,... 24 , n ( E ) 12
12
P ( 1 st even number)
25
This ticket is not replaced. Total is 24 . Out of which 11 are
even number.
nd 11
P ( 2 even number)
24
P( both even number) P( 1st even number & 2 nd even number)
12 11 11
25 24 50 43
AXIOMS OF PROBABILITY
3 . If P ( A B ) then P ( A B ) P ( A ) P ( B )
44
Laws of probability:
46
We know that the conditiona l probabilit y of event A
given that B has occurred is
47
Multiplication rule of probability for more than two events
Definition :
Two events A and B are said to be independen t,
if P ( B / A ) P ( B ), P ( A ) 0 and
P ( A / B ) P ( A ), P ( B ) 0
iii)w.k.t ( E F ) E F
P ( E F ) P (( E F ))
1 P( E F )
1 [ P ( E ) P ( F ) P( E F )]
1 P( E ) P( F ) P( E F )
50
Partition of a sample space
A set of events E , E , E ..... E is said to represent a partition
1 2 3 n
ii)S E E E .... .E
1 2 3 n
iii) P( E ) 0 for all i 1, 2, 3....n
i
51
Theorem of total probability
Let E , E , E ..... E
1 2 3 n
be a partition of the sample space S,
and suppose that each of the events E , E , E ..... E has 1 2 3 n
Proof :
( A E1 ) ( A E2 ) ( A E3 ) ......( A En )
Also A E and I
A E are respective ly the subsets
j
E and
i
E are disjoint,
j
for i j , therefore
A E I
and A E j
are also disjoint
for all i j, i , j 1 , 2 , 3 .... n . Thus,
P( A) P [( A E1 ) ( A E2 ) ( A E3 ) ......( A En )]
that there will be strike, 0.80 that the construction job will be completed
on time if there is no strike, and 0.32 that the construction job
will be completed on time if there is a strike. Determine the probability
that the construction job will be completed on time.
54
Solution.
Let A be the event that the construction job will be completed
on time, and B be the event that there will be a strike. We have
to find P(A).
We have P(B) 0.65, P(no strike) P( B ) 1 - P(B)
i)both white(eve nt E )1
ii)both green(even tE 2
)
5C2 5
P( E1 ) 3C 2 3
8C2 14 P( E2 )
8C 2 28
5C1 3C1 15
P( E3 )
8C2 28
56
After the balls are transferred from first urn to the second urn, the
second urn will contain
i)5W & 7G ii)3W & 9G iii)4W & 8G
Let A be the event of drawing a W ball from the second urn, Then
P ( A ) P ( E ) P ( A / E ) = 0.354
j 1 j j
57
Bayes’ Theorem
If E , E , E ..... E are n non empty events which
1 2 3 n
P (E )P (A / E ) j 1 j j
P ( A) P ( A)
P(E )P( A / E )
n ((by total prob.theo. )
i i
P(E )P( A / E )
j 1
j j
58
Note: The events E , E , E ..... E are called hypotheses .
1 2 3 n
of the hypothesis E i
5 0.49
P( A )P( X / A ) P( A )P( X / A )
1 1 2 2 3 5
0.4. 0.6.
5 12
5
P( A )P( X / A ) 0.6.
P( A / X ) 2 2
12
P( A )P( X / A ) P( A )P( X / A ) 0.51
2
1 1 2 2
3 5
0 .4 . 0 .6 .
5 12
60
2.In a bolt factory machines A , B & C manufactur es
respective ly 25%,35% ,40% of the total .Of their out put
5,4,2 % are defective bolts.A bolt is drawn from a day' s
production and found to be defective .What is the prob.
That it was manufactur ed by machine C .
Solution:
Let A, B & C be the events that the parts made by respective
machines.
P ( A) 0.25 P ( B ) 0 . 35 P (C ) 0.40
62