correlation
and regression
david hornby
Scattergrams are useful and the value of simple visual
inspection is most important in determining a trend between two
variables, whether or
not there is a
reasonable correlation between them. However, it is better if a more
objective and quantitative measure can be used to measure
the degree of
correlation. The "correlation coefficient" is a useful measure of
correlation and "regression analysis" allows the construction of
a regression equation
which can be used to predict future values and rents.
THE CORRELATION
COEFFICIENT (r) AND COEFFICIENT OF DETERMINATION (r2)
The purpose of
correlation analysis is to analyze the degree of relationship among the
values of 2 or more variables.
The measure of
correlation between 2 variables can be measured with "r2",
the "coefficient of determination":
Perfect positive
correlation: r2 = +1
No correlation: r2 =
0
Perfect negative
correlation: r2 = -1