M3 Part 2: Regression Analysis

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 21

M3 PART 2

REGRESSION ANALYSIS
Introduction

• Regression Analysis means the estimation or the prediction of the


unknown value of one variable from the known value of the other variable.
It is a statistical device used to study the relationship between two or more
variables that are related.
• Here there are two types of variables. The variable whose value is to be
predicted is called dependent variable and the variable which is used for
prediction is called independent variable.
• Simple and multiple regressions.
• On the basis of number of variables involved, regression is classified into
Simple and Multiple Regressions. When there are only 2 variables, it is
called simple regression. In multiple regression analysis, there are more
than 2 variables and we try to find out the effect of 2 or more independent
variables on one dependent variable.
• Linear & Non Linear Regression.
• On the basis of proportion of changes in the variables, the regression can
be classified into linear and non-linear regression. If the given bivariate
data are plotted on a graph, the points will more or less concentrate around
a curve, called “Curve of Regression”. If the regression curve is a straight
line, we say it is linear regression. If it is not a straight line, then it is
curved or non linear regression.

• We have seen that a scatter diagram is obtained, when the bivariate data are
plotted on a graph. We can draw straight lines through these points. But
that line which is very closer to the points is the line of best fit. This line is
also called regression line. It is a graphical technique to show the
functional relationship between the dependent and independent variables.
This line can be drawn using two methods such as free hand curve method
and the method of least squares. Of course, the latter is the more accurate
method.
• Regression equations are the equations of the
regression lines. Consider a general equation y = f(x).
Here “y” is the dependent variable and “x” is the
independent variable. This can be used only to
estimate the value of “y” for a given value of “x”.
• Similarly consider x = f(y). Here “x” is the dependent
variable and “y” is the independent variable. This can
be used only to estimate the value of “x” for a given
value of “y”.
• Thus there are 2 regression equations such as
Regression Equation of Y on X [y = f(x)] and
Regression Equation of X on Y [x = f(y)]. They are
not interchangeable.
A demo problem

• From the following data of the age of husband and the age of wife,
calculate the husband’s age when wife’s age is 16. Also find the wife’s age
when husband’s age is 40.
• Let “x” be the Husband’s age and “y” be the wife’s age.

H 36 23 27 28 28 29 30 31 33 35
age
W 29 18 20 22 27 21 29 27 29 28
age

• In regression analysis, don't divide by a common factor to


reduce the series as we did in correlation analysis. There is no
range for the values of regression coefficients.
• 1. Relation between Correlation coefficient and
regression coefficients.
• 2. How to get regression coefficients from regression
equations?
• 3. How to get the mean value of “x” and mean value
of “y” from regression equations.
• 4. How to identify the 2 regression equations, then
identifying the regression coefficients and thus to find
the correlation coefficient.
• 5. Difference between correlation and regression.
• Properties of regression lines.
• 1. The 2 lines intersect at (x̅ , ȳ) (x bar, y bar)
• 2. When r = 1, the 2 lines coincide.
• 3. When r = 0, the 2 lines are mutually perpendicular.
• Properties of Regression Coefficients.
• 1. Product of regression coefficients is the square of
correlation coefficient.
• 2. The sign of both the regression coefficients and that of
correlation coefficient is same.
• 3. When the correlation is perfect, the regression coefficients
will be reciprocal of each other.
• 4. Both the regression coefficients will not be greater than one.
That is, one of them can be greater than 1 or both can be less
than 1.
Distinction between Correlation and Regression

Correlation Regression

• We study degree of • We study the nature/pattern of


association/bonding between the relationship between the
variables. variables.
• The distinction between the • We need to clearly define which
dependent variable & the is the dependent variable and
independent variable is which is the independent
immaterial. variable.
• The purpose is not prediction • The basic purpose is predicting
but finding the bonding the value of dependent variable
between the variables & for a given value of the
expressing the same through a independent variable.
numerical value ranging from
-1 to +1.

You might also like