This preview shows page 1. Sign up to view the full content.
Unformatted text preview: on measures the nature of the association
Regression
nature
between the variables.
between
– Correlation measures the strength of the association
Correlation
between the variables.
between IESTomkins College 19 Regression/Correlation Analysis
Regression/Correlation
TYPES OF REGRESSION: Simple Linear Regression or BiVariate
Simple
Regression
Regression – Two variables:
» A dependent variable which is usually cost in our
dependent
analyses.
analyses.
» An independent variable or predictor variable which
An
is usually the measure of the volume of activity in
our analyses.
our
IESTomkins College 20 Regression/Correlation Analysis
Regression/Correlation A positive (+) value for b indicates that the dependent and
positive
the independent variables are positively correlated
(moving in the same direction). A negative () b value
indicates that the dependent and the independent variables
are moving in opposite directions. (Example, and increase
in interest rates is related to a decrease in construction
costs.)
costs.)
The a value indicates point where the regression line
The
crosses the “Y” axis. What does a negative? positive?
value for a indicate?
indicate? IESTomkins College 21 Regression/Correlation Analysis
Regression/Correlation The coefficient of determination measures the amount of
The
explained variance (i.e. Of the total variance of the
dependent variable about its mean (average) value, the
amount of that variance that can be explained by changes
in the volume of activity is measured by the coefficient of
determination.) For example, a coefficient of
determination of .84 or 84% means that 84% of the
dependent variable’s total variance can be explained by
changes in the volume of activity.
changes
Association vs. Causation  Regression/correlation analysis
Association
shows the degree of association between variables, but it
does not prove causation. IESTomkins College 22 Regression/Correlation Analysis
Regression/Correlation Standard Error of the Estimate:
Standard
Indicates the variability of the data used in
the regression calculations. The greater the
variability of the data, the less precision that
results from estimates made from the
regression data.
– Point estimates are made and confidence
Point
inter...
View Full
Document
 Spring '12
 na
 Accounting

Click to edit the document details