Linear Regression and Correlation

GE 112
Mathematics in the Modern World

Faculty, Math & Stat Department

College of Arts & Sciences
University of Southeastern Philippines

1
 Linear Regression
Introduction:

▪ We will examine the relationship between

quantitative variables x and y via a
mathematical equation.
▪ The motivation for using the technique:
- Forecast the value of a dependent variable (y)
from the value of one independent variable (x).
- Analyze the specific relationships between the
independent variable and the dependent
variable.
Linear Regression and
Correlation Page 2
 Linear Regression

The Model:
The model has a deterministic and a probabilistic
components.
House
Cost

Most lots sell

for $25,000

House size

Linear Regression and

Correlation Page 3
 Linear Regression
The Model:
However, house cost vary even among same
size houses! Since cost behave unpredictably,
we add a random component.

Most lots sell

for $25,000

House size

Linear Regression and

Correlation Page 4
 Linear Regression

The Model:
The first order linear model

y = 0 + 1x + 
0 and 1 are unknown population
parameters, therefore are estimated
𝑦 = dependent variable y from the data.

𝑥 = independent variable
0 = 𝑦-intercept
1 = slope of the line Rise  = Rise/Run
 = error variable 0 Run
x
Linear Regression and
Correlation Page 5
 Linear Regression
Estimating the Coefficients:

The estimates are determined by

- drawing a sample from the population of interest
- calculating sample statistics
- producing a straight line that cuts into the data

Question: What should be

y  considered a good line?



    
    


Linear Regression and

Correlation Page 6
 Linear Regression

The Least Squares (Regression) Line

A good line is one that minimizes

the sum of squared differences between the
points and the line.

Linear Regression and

Correlation Page 7
 Linear Regression

Formula:

To calculate the estimates of the slope and intercept

of the least squares line , use the formulas:
SS xy
b1 =
SS xx
b0 = y − b1 x

where SS xy =  xi yi −
(  x )(  y )
i i

n
(  x)
2

SS xx =  x − = (n − 1) sx2
2 i
i
n
Linear Regression and
Correlation Page 8
 Linear Regression
The regression equation that estimates the equation of the
first order linear model is:

ŷ = b0 + b1 x
Note:

1. b0 is the intercept of the equation. It is not

interpreted if the possible values of the
independent variable x does not contain 0.
2. b1 is the slope of the equation. If b1 is positive
(negative), then the estimate value in y increases
(decreases) with an amount equal to b1 in every
one unit increase in x

Linear Regression and

Correlation Page 9
 Linear Regression
Example.

The grades of a class of 9 students on a midterm report (x)

and on the final examination (y) are as follows:
x 77 50 71 72 81 94 96 99 67
y 82 66 78 34 47 85 99 99 68

a) Estimate the linear regression line.

b) Estimate the final examination grade of a student who

received a grade of 85 on the midterm report.

c) Interpret b1.

Linear Regression and

Correlation Page 10
 Linear Regression
Solution.

a) The estimated linear regression line is

𝑦ො = 12.06 + 0.78𝑥

b) The estimate final examination grade of a

student who received a grade of 85 on the
midterm report is

𝑦ො = 12.06 + 0.78 85 = 78.36

Linear Regression and

Correlation Page 11
 Linear Regression

c) For every 1 point increase in midterm

report (x), the final examination (y) will
increase by about 0.78 points.

Linear Regression and

Correlation Page 12
 Correlation

Linear Correlation Coefficient (r)

It determines the strength of a linear

relationship between two variables x and y
and is computed using the formula

Linear Regression and

Correlation Page 13
 Correlation
Note:

1. If r is positive, the relationship between the variables has a positive

correlation. In this case, if one variable increases, the other
variable also tends to increase.

2. If r is negative, the linear relationship between the variables has a

negative correlation. In this case, if one variable increases, the
other variable tends to decrease.

3. The closer |r| is to 1, the stronger the linear relationship between the
variables.

4. If r is close to zero it does not mean always that there is low or no

relationship between variables. r measures only the linear
relationship so it is always best to obtain a scatter plot also of your
data to maximize the information it can give.

Linear Regression and

Correlation Page 14
 Correlation

Types of linear correlation

Linear Regression and

Correlation Page 15
 Correlation

Types of linear correlation

Linear Regression and

Correlation Page 16
 Correlation
Example.

The grades of a class of 9 students on a midterm

report (x) and on the final examination (y) are as
follows:
x 77 50 71 72 81 94 96 99 67
y 82 66 78 34 47 85 99 99 68

a) Obtain a scatter plot of the data.

b) Compute r and assess the relationship between

midterm report and final examination.

Linear Regression and

Correlation Page 17
 Correlation
Solution.

a) Scatter plot between midterm report and final examination

Linear Regression and

Correlation Page 18
 Correlation

b) The value of r = 0.56. It indicates that there

is a positive (and moderately strong)
relationship between midterm report and final
examination. Also it indicates that if midterm
report increases then final examination will
tend to increase also.

Linear Regression and

Correlation Page 19
 Using Calculator for Computations

Note: The video uses A for intercept (while we

use b0) and B for the slope of the regression
equation (we use b1). The illustration in this
video will only be applicable to the model of
the calculator (or similar models) used by the
author and might not work in your calculator.
So check it out!

Linear Regression and

Correlation Page 20
 Using Calculator for Computations

Linear Regression and

Correlation Page 21
 References

Aufmann, R., Lockwood, J., Nation, R. and

Clegg, D. (2013). Mathematical Excursions
(3rd Edition). Brooks/Coole: Cengage
Learning

Walpole, R., Myers, R., Myers, S. and Ye, K. (2007).

Probability and Statistics for Engineers and
Scientists (8th Edition). Upper Saddle River,
N.I.: Pearson Prentice Hall
Linear Regression and
Correlation Page 22