BITS Pilani

Pilani Campus

Course No: MTH F113

Probability and Statistics

BITS Pilani, Pilani Campus
 BITS Pilani
Pilani Campus

Section 5.3 and Chapter 11: Simple Linear

Regression and Correlation
Sumanta Pasari
(sumanta.pasari@pilani.bits-pilani.ac.in) BITS Pilani, Pilani Campus
 Recall: 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.

3 BITS Pilani, Pilani Campus

 Recall: 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 
4 BITS Pilani, Pilani Campus
 Recall: Covariance Properties

(i) If 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 

5 BITS Pilani, Pilani Campus

 Correlation
Correlation is a normalized covariance. It provides a linear relationship between X and Y .
If X and Y show similar behaviour, then the correlation is +ve, else -ve. Let X , Y be RVs
with means  X , Y , and variance  X2 ,  Y2 , respectively. Then
Cov  X , Y 
 XY 
 XY
Remarks:
(i)  1   XY  1
(ii) If two RVs X and Y are independent, then they are uncorrelated   XY  0  .
(iii) However,  XY  0 does not imply that X and Y are independent (WHY?).
(iv)  XY  YX and  XX  1
a
(v)   aX , Y    XY and   X  a, Y  b    XY
a
ac
(vi)   aX  b, cY  d    XY
a c
6 BITS Pilani, Pilani Campus
 Problem Solving
ac
Ex. 11.1. Show that   aX  b, cY  d    XY ; a  0, c  0
a c
Cov  aX  b, cY  d 
Sol.   aX  b, cY  d  
  aX  b    cY  d 
E  aX  b  cY  d    E  aX  b  E  cY  d 

  aX  b    cY  d 
E  acXY  adX  bcY  bd    aE  X   b  cE Y   d 

a   X  c  Y 
acE  XY   adE  X   bcE Y   bd  acE  X  E Y   adE  X   bcE Y   bd

a c   X   Y 
ac E  XY   E  X  E Y  ac
   XY  Proved 
a c   X   Y  a c
7 BITS Pilani, Pilani Campus
 Problem Solving
kxy; 0  x  1, 0  y  x
HW. 11.1. Let f  x, y   
 0 ; e.w.
(i) Find k , E  XY  (ii) Var  X  , Var Y  (iii) Cov  X , Y  ,  XY
k  3x  y  ; 1  x  3, 0  y  2
HW11.2. If f  x, y    is a pdf,
 0 ; e.w.
Find (i) k (ii) Cov  X , Y  ,  XY .
k sin x sin y; 0  x   2, 0  y   2
HW11.3. If f  x, y    is a pdf,
 0 ; e.w.
Find (i) k (ii) Cov  X , Y  ,  XY .
k  6  x  y  ; 0  x  2, 2  y  4
HW11.4. If f  x, y    is a pdf,
 0 ; e.w.
(i) Find k and (ii) Cov  X , Y  ,  XY .
8 BITS Pilani, Pilani Campus
 Problem Solving
HW. 11.5. 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 Y  , E  XY 
(iv) Cov  X , Y  ,  XY .

HW11.6. Show that the linear combinations aX  bY and cX  dY

of the random variables X and Y are uncorrelated if
ac x2   ad  bc   x y  bd y2  0.

9 BITS Pilani, Pilani Campus

 Section 11.6: Correlation
So, we are trying to measure the linear relation
between two random variables X and Y.
The theoretical parameter used to measure the
linear relationship between X and Y is the
Pearson coefficient . This parameter is defined
by
Cov ( X , Y )

(Var X )(VarY )

BITS Pilani, Pilani Campus

 Correlation

The parameter  assumes values between -1

and 1 inclusive. Values of -1 and 1 indicate
perfect positive and negative relationships
respectively. A value of 0 indicates no linear
relationship. When this occurs, we say that
X and Y are uncorrelated.

BITS Pilani, Pilani Campus

 Correlation
Previously, we found that the theoretical
value of  based on knowledge of the joint
density function for X and Y. But these are
seldom known in practice. So our job is to
estimate  based on set of observations
{(xi,yi):i=1,2,…n} on the random variable
(X,Y). We must use Var X, Var Y and Cov
(X,Y) for this purpose.

BITS Pilani, Pilani Campus

 Correlation

 n
Var X   ( Xi  X ) / n  Sxx / n
2

i 1
 n
Var Y   (Yi  Y ) / n  Syy / n
2

i 1

To estimate Cov(X,Y), note that

Cov( X , Y )  E[( X   )(Y   )]
X Y

BITS Pilani, Pilani Campus

 Correlation
We estimate Cov(X,Y) by averaging products
analogous to that on the right hand side of the
previous equation.
 n
Cov( X , Y )   ( Xi  X )(Yi  Y ) / n  Sxy / n
i 1

When we combine these estimators, the estimator

for  is given by
 Sxy
R
SxxSyy
BITS Pilani, Pilani Campus
 Computational Formula for r
The computational formula for r (the estimated
Pearson correlation coefficient) is
n xy  x y
ˆ est  r 
[n x 2  ( x) 2 ][n y 2  ( y ) 2 ]
or ,
n
 n  n 
n xi yi    xi   yi 
rXY  i 1  i 1  i 1 
2 2
n
  n n
 n

n x    xi 
2
i n y    yi 
2
i
i 1  i 1  i 1  i 1 
BITS Pilani, Pilani Campus
Correlation Coefficient

16 BITS Pilani, Pilani Campus

 Problem Solving

Ex. 11.2 (Ex. 47, page 435) Pesticides used in food

production can be found in food consumed by humans.
A study focusing on chickens exposed to malaoxon was
conducted. The chickens were also exposed to a liver
enzyme inducer to determine whether liver
detoxification of the pesticide is affected. The following
data were reported as a percentage of normal pesticide
detoxification (y) and percentage of normal liver enzyme
levels (x)

BITS Pilani, Pilani Campus

 Problem Solving
Enzyme level (x) Detoxification level (y)
95 108
110 126
118 102
124 121
145 118
140 155
185 158
190 178
205 159
222 184
BITS Pilani, Pilani Campus
 Problem Solving

47.a) Plot a scattergram of the data.

b) Estimate , the correlation between X and Y.

BITS Pilani, Pilani Campus

 Problem Solving
x y x2 y2 xy
95 108 9025 11664 10260
110 126 12100 15876 13860
118 102 13924 10404 12036
124 121 15376 14641 15004
145 118 21025 13924 17110
140 155 19600 24025 21700
185 158 34225 24964 29230
190 178 36100 31684 33820
205 159 42025 25281 32595
222 184 49284 33856 40848
S1534 S1409 S252684 S206319 S226463
BITS Pilani, Pilani Campus
 Problem Solving

n = 10, Sx = 1534, Sy  1409, Sx2 = 252684,

Sy2 = 206319, Sxy = 226463

 n xy  x  y
 est  r 
[n x 2  ( x) 2 ][n y 2  ( y ) 2 ]
10  226463  1534  1409

[10  252684  (1534) 2 ][10  206319  (1409) 2 ]
 0.8874

BITS Pilani, Pilani Campus

 Problem Solving
Ex. 11.3 (Ex. 7, page 427) The relationship between energy
consumption and household income was studied, yielding the
following data on household income X ( in units of $1000/year)
and energy consumption Y (in units of 108 Btu/Year)
Energy Consumption (y) Household Income (x)
1.8 20.0
3.0 30.5
4.8 40.0
5.0 55.1
6.5 60.3
7.0 74.9
9.0 88.4
9.1 95.2
BITS Pilani, Pilani Campus
Househol Energy
 xi  x   yi  y 
d Income Consumption 2 2
(x) (y)  xi  x  yi  y 
20 1.8 1447.8 15.8 151.24875
30.5 3 759 7.701 76.45125
40 4.8 325.8 0.951 17.59875
55.1 5 8.7025 0.601 2.28625
60.3 6.5 5.0625 0.526 1.63125
74.9 7 283.92 1.501 20.64125
88.4 9 921.12 10.4 97.87875
95.2 9.1 1380.1 11.06 123.52375
S 464.4 S 46.2 S5131.5 S48.54 S 491.26

BITS Pilani, Pilani Campus

Problem Solving
n
sxx   ( xi  x ) 2  5131.54
i 1
n
s yy   ( yi  y ) 2  48.54
i 1
n
sxy   ( xi  x )( yi  y )  491.26
i 1

sxy
r  0.9843
sxx s yy

BITS Pilani, Pilani Campus

 Computation of Correlation
Coefficient
HW 11.7: Data on heart disease and cigarette smoking Country Cigarettes CHD
1 11 26
in 21 developed countries (Landwehr and Watkins, 2 9 21
1987) are provided. Find the covariance and 3 9 24
4 9 21
correlation coefficient between cigarette consumption 5 8 19
(per adult per day) and coronary heart disease (CHD) 6 8 13
7 8 19
mortality per 10,000. (Answer: 10.60, 0.71) 8 6 11
9 6 23
10 5 15
Hint: 11 5 13
12 5 4
n
 n  n  13 5 18
n xi yi    xi   yi  14 5 12

rXY  i 1  i 1  i 1  15 5 3
16 4 11
2 2
n
 n
 n
 n
 17 4 15
n xi2    xi  n yi2    yi  18 4 6
i 1  i 1  i 1  i 1  19 3 13
20 3 4
21 3 14

BAZC413: Introduction to Statistical Methods 25 BITS Pilani, Pilani Campus

 Problem Solving
HW. 11.8. An engineer found
that by including small Amount of Life (y)
amount of compound Additive (x)
rechargeable batteries for 5 10
portable computers, she
could extend their lifetimes. 15 18
She experimented with 25 20
different amounts of the
additive and the data are as 35 25
below.
45 32
Obtain a point estimate of
correlation. 50 45
BITS Pilani, Pilani Campus
 Problem Solving

 x  175,  y  150,  x 2
 6625,   4498,  xy  5385
y 2

n xy  x y
ˆ est  r 
[n x 2  ( x) 2 ][n y 2  ( y ) 2 ]
6  5385  175  150

[6  6625  (175) 2 ][6  4498  (150)2 ]
 0.9469.

BITS Pilani, Pilani Campus

 Problem Solving

HW. 11.9. Based on the given x y

data,
a) Plot a scatter diagram of the 4 8
data.
2 12
b) Estimate r, the correlation 10 4
between X and Y
5 10
8 2

BITS Pilani, Pilani Campus

 Problem Solving

x y x2 y2 xy
4 8
16 64 32
2 12
4 144 24
10 4
100 16 40
5 10
25 100 50
8 2
64 4 16

S29 S36 S209 S328 S162

BITS Pilani, Pilani Campus
 Problem Solving

n xy  x y
ˆ est  r 
[n x 2  ( x) 2 ][n y 2  ( y ) 2 ]
5  162  29  36

[5  209  (29) 2 ][5  328  (36) 2 ]
 0.883

BITS Pilani, Pilani Campus

 BITS Pilani
Pilani Campus

Section 11.1: Simple Linear

Regression

BITS Pilani, Pilani Campus

 Let’s Observe…
(1) Relationship between degrees Fahrenheit and degrees Celsius:
9
F  C  32  deterministic 
5
(2) Circumference    diameter  C  2 r  deterministic 
(3) Height and weight of students (is there a perfect relationship?)
(4) Driving speed and gas mileage (is there a deterministic relation?)
(5) Fertilizer and crop yield (production)
(6) Drug dosage and time to get cured
(7) Income and expenditure of a group of persons
(8) Sunshine hours/temperature and icecream sale
(9) Age of car and its sale price
32 BITS Pilani, Pilani Campus
Let’s Observe…linear or non-linear?

33 BITS Pilani, Pilani Campus

 Purpose: Linear Regression
For the following observed data, how can we get the
“best-fit” line? What would be the equation?

34 BITS Pilani, Pilani Campus

 How to Formulate?
(1) Recall that we studied conditional mean, say Y x .
What does Y x mean?
(2) Suppose, X is NOT a RV, rather a mathematical variable.
e.g., let X : depth of water, Y : the water temperature.
Then can we model the water temperature Y , as a function of X?
(3) Thus, aren't we dealing with a conditional variable Y x ?
(4)This Y x will have a mean Y x (a function of x).
(5) Can I express  linearly  this as Y x = 0  1 x ?
(Linear Curve of Regression of Y on X )
35 BITS Pilani, Pilani Campus
 Simple Linear Regression
Simple linear regression (regression means ‘act of going
back’, ‘return’, or ‘reversion’) is a statistical method that
allows us to summarize and study relationships between two
continuous (quantitative) variables:
• One variable, denoted by X, is regarded as
the predictor, explanatory, or independent variable.
• The other variable, denoted by Y, is regarded as
the response, outcome, or dependent variable.

Why is it called “Simple Linear Regression” model? What is a

model? Why simple? Why linear? What is regression?
36 BITS Pilani, Pilani Campus
 What is a Model? Role Model?

Representation of some phenomenon.

However, these are non-math/non-stats model.. then??
37 BITS Pilani, Pilani Campus
 Math/Stats Model
1. Often, they describe relationship between variables
2. Could be deterministic, or probabilistic (stochastic)

Probabilistic
Models

Regression Correlation Other

Models Models Models

38 BITS Pilani, Pilani Campus

 Regression Model

1 Explanatory Regression 2+ Explanatory

Variable Models Variables

Simple Multiple

Non- Non-
Linear Linear
Linear Linear

39 BITS Pilani, Pilani Campus

11.1 MODEL AND PARAMETER
ESTIMATION

In simple linear regression model,

Y x = 0  1 x
 0 denotes the intercept and 1 is the slope
of the simple linear regression line.

X: Predictor, Explanatory, or Independent variable

Y: Response, Outcome, or Dependent variable

BITS Pilani, Pilani Campus

 Formulation: SLEM
(Simple Linear Regression Model)
Let x1 , x2 , ……. xn be values of X for which observations
are made. These points are assumed to be measured
without error.

Note that x1, …, xn could be preselected by the

experimenter (controlled study) or they are observed at
random (observational study).

In any case, we have random sample as (x1, Y|x1),

(x2,Y|x2),….. (xn,Y|xn). Recall that these RVs Y|xi vary
about their mean value. Let E Y | x  
i i Y xi
BITS Pilani, Pilani Campus
 Formulation: SLEM
Solving, Y | xi  Ei  Y xi

In this expression, we have assumed that the

random difference Ei has a mean zero. Since we
are assuming a linear regression, we have
Y | xi  0  1 xi  Ei
It is customary to drop the conditional notation
and denote Y|xi by Yi. Hence
Yi  0  1 xi  Ei
BITS Pilani, Pilani Campus
 Formulation: SLEM
Our data: (xi , yi) i= 1, 2, … n, where xi denotes
the observed value of the variable X and yi is the
observation for the random variable Y. This above
idea is mathematically expressed as
y i =  o +  1 x i +e i

In this equation , ei denotes the realization of the

random variable Ei when Yi takes the value yi .

BITS Pilani, Pilani Campus

 Formulation: SLEM
In a regression study, it is useful to plot the data
points in xy-plane. Such a plot is called the
scattergram (scatter diagram). We do not expect the
points to lie exactly on a straight line. However if
linear regression is applicable, then they should
exhibit a linear trend.

BITS Pilani, Pilani Campus

 Formulation: SLEM
Note that since we do not know the true values of
0 and 1 (WHY??), we shall not know the true
value of ei (the vertical distance from (xi , yi) to
the true regression line).

Letting b0 and b1 denote the estimates of o and

1 respectively, the estimated line of regression
takes the form,
ˆY x = b0  b1 x
BITS Pilani, Pilani Campus
 Formulation: SLEM
Just as the data points do not all lie on the theoretical
line of regression, they also do not all lie on this
estimated regression line. If we let ei denote the
vertical distance from a point (xi , yi ) to the estimated
regression line, then each data point satisfies the
equation
yi  b0  b1 xi  ei
This term ei is termed as residual.
What is the difference between ei and ei ?
BITS Pilani, Pilani Campus
Population and Sample SLEM

BITS Pilani, Pilani Campus

ei and ei

BITS Pilani, Pilani Campus

 Least –Squares Estimation
The parameters 0 and 1 are determined by
method of least squares. This works best because
we wish to pick the one that that best fits the
data. We choose b0 and b1 such that we minimize
the sum of the squares of the residuals.
The sum of the squares of the residual errors
about the estimated regression line is given by
n n
SSE   ei2   ( yi  b0  b1xi ) 2
i 1 i 1
BITS Pilani, Pilani Campus
 Least –Squares Estimation

Differentiating this with respect to b0 and b1 we obtain,

SSE n
 2 ( yi  b0  b1 xi )
b0 i 1

SSE n
 2 ( yi  b0  b1xi )xi
b1 i 1

BITS Pilani, Pilani Campus

 Least –Squares Estimation

We set these partial derivatives as zero and use the rules

of summation to obtain the equations
n n
nb0  b1  xi  yi
i 1 i 1
n n n
b0  xi  b1  x  xi yi
2
i
i 1 i 1 i 1

These equations are called the normal equations.

They can be solved easily for b0 and b1 , that is, the
estimates for 0 and 1 , respectively.

BITS Pilani, Pilani Campus

Least-Squares Estimates for β0 and β1

n
 n
n

n xi yi    xi )    yi 
b1  i 1  i 1   i 1 
2
n
 n

n xi    xi ) 
2

i 1  i 1 
b0  y  b1 x
BITS Pilani, Pilani Campus
Least-Squares Estimation

53 BITS Pilani, Pilani Campus

Alternative Approach
2
 
n
n  i   i 
x 2
 x
S xx   ( xi  x ) 2  ;
i 1 n
n n xi yi    xi   yi 
S xy   ( xi  x )( yi  y )  ;
i 1 n
n y    yi 
2 2
n
S yy   ( yi  y ) 2 i
.
i 1 n
Then
n
SSE    yi  b0  b1 xi     (Yi  y )  b1 ( xi  x ) 
2 2

i 1

 S yy  2b1 S xy  b12 S xx
BITS Pilani, Pilani Campus
Alternative Approach

To minimize this, equate derivative wrt b1 to 0 ,

S xy
2 S xy  2b1 S xx  0  b1  .
S xx
Now b0  y  b1 x.

Note:
A combined question may be asked to estimate the
regression line and then to get the point estimate of
correlation. The same tabulation will work for both.
BITS Pilani, Pilani Campus
Need Computation Table

56 BITS Pilani, Pilani Campus

 Problem Solving
Ex. 11.4: In the following table, x is the tensile force applied
to a steel specimen in thousand of founds and y is the
resulting elongation in thousands of an inch:

x 1 2 3 4 5 6
y 14 33 40 63 76 85
(a)Graph the data to verify that it is reasonable to assume that
the regression of Y on X is linear
(b) Find the equation of the least-squares line and use it to
predict the elongation when the tensile force is 3.5
thousand pounds.
BITS Pilani, Pilani Campus
 Problem Solving
(a) Do by yourself
(b) Since n=6, we need to have:
n n n n n

 xi  21,
i 1
 i  91,
x 2

i 1
 yi  311,
i 1
 xi yi  1342,
i 1
 i  19855,
y 2

i 1

Estimated Regression Line:

ˆY x  b0  b1 x
 yˆ  1.133  14.486 x  interpretation of results? 
Therefore, when the tensile force is 3.5, y = 51.83
BITS Pilani, Pilani Campus
 Problem Solving
Ex. 11.5 (Ex. 2, Page 427) For the following data sets,
plot a scattergram and subjectively state whether it
appears that a linear regression will (i) fit the data well
(ii) give only a fair fit (iii) fit the data poorly.
x 5 15 25 35 45 50

Y 10 18 20 25 32 45

Ex. 6. Now, estimate 0 and 1 . Find the residuals in

each case and verify that apart from round-off error, the
residuals sum to 0.
BITS Pilani, Pilani Campus
 Problem Solving
n
 n  n 
n  xi yi    xi )    yi 
b1  i 1  i 1   i 1   0.66
2
n
 n

n  xi    xi ) 
2

i 1  i 1 
b0  25  0.66(29.2)  5.7
yˆ  5.7  0.66 x
yi 10 18 20 25 32 45
estimated y 8.95 15.59 22.23 28.87 35.52 38.84
ei 1.05 2.41 -2.23 -3.87 -3.52 6.16
e 0
i
BITS Pilani, Pilani Campus
 Problem Solving

Ex. 11.6 (Ex. 7, Page 427) The relationship

between energy consumption and
household income was studied, yielding the
following data on household income X (in
units of $1000/year) and energy
consumption Y (in units of 108 Btu/Year)

BITS Pilani, Pilani Campus

Problem Solving
Energy Consumption (y) Household Income (x)

1.8 20.0
3.0 30.5
4.8 40.0
5.0 55.1
6.5 60.3
7.0 74.9
9.0 88.4
9.1 95.2
BITS Pilani, Pilani Campus
 Problem Solving

• Plot a scatter gram of these data.

• Estimate the linear regression equation
Y|x= 0 + 1x.
• If x = 50 ( household income of $50,000),
estimate the average energy consumed for
households of this income. What would
your estimate be for a single household?

BITS Pilani, Pilani Campus

 Problem Solving
Solution: (a).
(b) Since n=8, we need to have:
n n n

 xi  464.4,
i 1
 i  32089.96,
x 2

i 1
y i 1
i  46.2,

n n

 xi yi  3173.17,
i 1
 i  19855,
y 2

i 1

Estimated Regression Line: yˆ  0.2177  0.0957 x

Therefore, when x = 50, y = 5.004

BITS Pilani, Pilani Campus
 n n
sxx   xi2  ( xi ) 2 / n  5131.54
i 1 i 1
n n n
sxy   xi yi  ( xi )( yi ) / n  491.26
i 1 i 1 i 1

sxy  
b1   0.095733, b0  y  b1 x  0.2177
sxx

Therefore, when x = 50, we have:

y  0.2177  (0.95733)(50)  5.004
BITS Pilani, Pilani Campus
Household Energy
Income (x) Consumpti
on (y)
(xi-x)2 (yi-y)2 (xi-x)(yi-y)
20 1.8 1447.8 15.8 151.24875
30.5 3 759 7.701 76.45125
40 4.8 325.8 0.951 17.59875
55.1 5 8.7025 0.601 2.28625
60.3 6.5 5.0625 0.526 1.63125
74.9 7 283.92 1.501 20.64125
88.4 9 921.12 10.4 97.87875
95.2 9.1 1380.1 11.06 123.52375
S 464.4 S 46.2 S5131.5 S48.54 S 491.26
BITS Pilani, Pilani Campus
Problem Solving

n
s xx   ( xi  x )  5131.54
2

i 1
n
s xy   ( xi  x )( yi  y )  491.26
i 1
s xy
b1   0.095733, b0  y b1 x  0.2177
s xx

BITS Pilani, Pilani Campus

 Problem Solving

HW 11.10 (Ex. 8.) Consider data in exercise 7.

a)Write the normal equations for these data.
b)Solve the normal equations for b0 and b1. Verify
that your results are the same as those you
obtained in part (b) of exercise 7.

BITS Pilani, Pilani Campus

Problem Solving
normal equations are:
n n
nb 0  b1 xi  yi
i 1 i 1
n n n
b 0  xi  b1 xi  xiyi 2

i 1 i 1 i 1

8b 0  464.4b1  46.2
464.4b0  32089.96b1  3173.17

BITS Pilani, Pilani Campus

 Problem Solving
HW 11.11: The city’s transportation department is interested in studying
the relationship between the temperature and the number of
passengers that use public transportation. The manager recorded the
temperature at the beginning of the hour, and then had a bus driver
record the number of passengers that boarded the bus throughout the
hour. Their findings are listed below.

temperature 42 37 46 30 50 43 43 46 46 49

passenger 173 149 185 123 201 174 175 188 186 198

(a) Find an estimated regression line of Y on X.

(b) Obtain a point estimation of correlation coefficient.
BITS Pilani, Pilani Campus