Scribd
Upload
54 views22 pages

Regression Analysis: Prof. Prema Muthuswamy KCT, Coimbatore

The document discusses linear regression analysis. [1] Linear regression is used to predict the value of a dependent variable from one or more independent variables and explore the relationship between variables. [2] Simple linear regression uses one independent variable while multiple linear regression uses two or more independent variables. [3] Examples show how linear regression finds the best-fit line that minimizes the difference between predicted and actual dependent variable values.

Uploaded by

sam sushmitha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
54 views22 pages

Regression Analysis: Prof. Prema Muthuswamy KCT, Coimbatore

The document discusses linear regression analysis. [1] Linear regression is used to predict the value of a dependent variable from one or more independent variables and explore the relationship between variables. [2] Simple linear regression uses one independent variable while multiple linear regression uses two or more independent variables. [3] Examples show how linear regression finds the best-fit line that minimizes the difference between predicted and actual dependent variable values.

Uploaded by

sam sushmitha
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 22

Regression Analysis

Prof. Prema Muthuswamy


KCT, Coimbatore
Purpose of Regression
• To predict the value of a
dependent variable from the
values of at least one explanatory
or independent variable
(predictors)
• To explore the nature of the
relationship among these
variables
Regression Analysis
• Majority of regression analyses
use a linear regression model

• Dependent variable expressed in


terms of linear combinations of
independent variables
Relationship

To PREDICT, there should be a


statistical relationship between
variables ( variables that are NOT
perfectly related)
Statistical Relationship
Examples
• Height & Weight
• Driving speed & fuel mileage
• Running speed and oxygen intake

To check if relationship exists between variables,


run Correlation test
Types of Linear Regression

• Simple Linear Regression


• Multiple Linear regression
Simple Linear Regression
• One predictor variable
(Independent Variable)
• Predictor can be continuous,
dichotomous or categorical
• Only one response (dependent
variable), and it is continuous.
Simple Linear Regression -
Equation

This is of the form: Y = C + mx


(C is a constant, m is the slope of the line)
Simple Linear Regression

Slope dy/dx
(b1)

Intercept
(Constant
bo)
Linear Regression –
Examples
Simple Linear Regression
– Example 1 (Data)
Height in Weight in
Age in months
inches pounds
191 62.5 112.5
160 62 94.5
157 64.5 123.5
191 65.3 107
141 61.8 85
185 63.3 101
149 64.3 110.5
169 62.3 99.5
173 62.8 102.5
….. ….. …..
Simple Linear Regression
Example 1 – Model Significance
62% of the variation
in Weight is
explained by Height

Regression Model is Significant & Prediction is possible


Simple Linear Regression
Example 1- Parameter Estimates

Predicted equation:
Weight = - 127.91 + (3.729 X Height)
Simple Linear Regression
Example 1 – Regression Plot
Simple Linear Regression Example 1 –
Regression Plot BEST FIT LINE

Actual Value
Regression
Line

Error/ Predicted Value


Deviation
Simple Linear Regression Example 1 –
Regression Plot BEST FIT LINE
Age in Height in Weight in Predicted
Error (Error)2
months inches pounds Weight
191 62.5 112.5 105.2 7.3 54
160 62 94.5 103.3 -8.8 77
157 64.5 123.5 112.6 10.9 119
191 65.3 107 115.6 -8.6 74
185 63.3 101 108.1 -7.1 51
149 64.3 110.5 111.9 -1.4 2
169 62.3 99.5 104.4 -4.9 24
173 62.8 102.5 106.3 -3.8 14
150 61.3 94 100.7 -6.7 45
144 59.5 93.5 94.0 -0.5 0
….. …… ……. ……. ….. …….

Sum of Square of the Error term = 4554


Simple Linear Regression Example 1 –
Regression Plot BEST FIT LINE

• Sum of Square of the Error term = 4554

• Best FIT line is the one that has this


sum as small as possible
(Least Squares Criterion)

• No other line will have the this


SUM = or < 4554
Simple Linear Regression
Example 1 – Actual by Predicted Plot
Multiple Linear Regression
– Example 2 (Same Data)
Height in Weight in
Age in months
inches pounds
191 62.5 112.5
160 62 94.5
157 64.5 123.5
191 65.3 107
141 61.8 85
185 63.3 101
149 64.3 110.5
169 62.3 99.5
173 62.8 102.5
….. ….. …..
Multiple Linear Regression
Example 2 – Model Significance
65% of the variation
in Weight is
explained by Height
& Age

Regression Model is Significant & Prediction is possible


Multiple Linear Regression
Example 2- Parameter Estimates

Predicted equation:
Weight = - 121.26 + (3.06 X Height) + (0.2067 X Age)
Multiple Linear Regression
Example 2 – Actual by Predicted Plot

You might also like