LINEAR REGRESSION

ANALYSIS
REGRESSION ANALYSIS
• It is a method of analyzing the variability of a response variable by using
information available on one or more regressor variables.
• It seeks to answer the question: What are the expected changes in the response
variable as a result of changes (observed or induced) in the regressor
variable(s)?
• A response variable (or dependent variable) is a variable whose values we wish
to characterize.
• A regressor variable (or independent variable, also known as predicting
variable) is a variable whose effects on the response variable we wish to study
REGRESSION ANALYSIS
• The most commonly used form of regression is linear regression, and the most
common type of linear regression is called ordinary least squares regression.
ASSUMPTIONS OF LINEAR REGRESSION
In theory, there are several important assumptions that must be satisfied if linear regression is
to be used. These are:
• Both the independent (X) and the dependent (Y) variables are measured at the interval or
ratio level.
• The relationship between the independent (X) and the dependent (Y) variables is linear.
• Errors in prediction of the value of Y are distributed in a way that approaches the normal
curve.
• Errors in prediction of the value of Y are all independent of one another.
• The distribution of the errors in prediction of the value of Y is constant regardless of the
value of X.
THE LEAST SQUARES LINEAR REGRESSION EQUATION
•
THE LEAST SQUARES LINEAR REGRESSION EQUATION
•
 EXAMPLE
Data on selected families of students in PQR University and corresponding income;
Number of members in the family (X)
Monthly family income in thousand of pesos (Y)

Student X Y
1 3 64
2 8 150
3 5 101
4 2 36
5 2 42
6 4 75
7 6 124
 EXAMPLE
1. What is the simple linear regression equation?

2. What is the effect on Y if the number of family member is increased by 1 person?

3. What is the estimated value of Y when X = 10? What is the interpretation?

Solution:

Student X Y
1 3 64
2 8 150
3 5 101
4 2 36
5 2 42
6 4 75
7 6 124
30 592
Solution:

Student X Y
1 3 64 192 9
2 8 150 1200 64
3 5 101 505 25
4 2 36 72 4
5 2 42 84 4
6 4 75 300 16
7 6 124 744 36
30 592 3097 158
•
•
 ANSWER
•
 EXAMPLE
A teacher would like to determine whether the students score in Algebra has an effect on their scores in Calculus.

Student Algebra (X) Calculus (Y)

1 17 73
2 21 66
3 11 64
4 16 61
5 15 70
6 11 71
7 24 90
8 27 68
9 19 84
10 8 52
 EXAMPLE
1. What is the simple linear regression equation?

2. What is the effect on Y if the score in Algebra is increased by 1 point?

3. What is the estimated value of Y when X = 25? What is the interpretation?

Solution:
Student X Y
1 17 73
2 21 66
3 11 64
4 16 61
5 15 70
6 11 71
7 24 90
8 27 68
9 19 84
10 8 52
169 699
Solution:
Student X Y
1 17 73 1241 289
2 21 66 1386 441
3 11 64 704 121
4 16 61 976 256
5 15 70 1050 225
6 11 71 781 121
7 24 90 2160 576
8 27 68 1836 729
9 19 84 1596 361
10 8 52 416 64
169 699 12146 3183
•
•
 EXAMPLE
•