FINANCE Corporate financial policy and R and D Management

(5.14)

To test whether the model is a useful model, an F-test is performed
where:

H 0 = α= β= 0

HA= α/β= 0

(5.15)

As the calculated F-value exceeds the critical F-value with (1, N– 2) de-
grees of freedom of 5.99 at the .05 level of significance, the null hypothesis
must be rejected. The 95 percent confidence level limit of prediction can be
found in terms of the dependent variable value:

(5.16)

Multiple Regression

Business and economic analysts generally are interested in the way in
which a dependent variable is related to more than one independent vari-
able. The overall purpose of this chapter is to describe the nature and ap-
plication of multiple regression and correlation techniques, the methods
that are used when there is more than one independent variable. Among
the specific objectives are:

 To show how one can calculate and interpret the intercept and slopes

of a multiple regression (two or more independent variables).

 To define the multiple coefficient of determination and indicate how it

can be computed and used.

(ˆ ˆ )/

()

abX ta S

NX X

NM

e

XX

++

−

++

02

0

2

1

F

YX

eN

i

N

i

N

i

= N

÷−

÷−

= =

=

∑ ∑

∑

22

1

2

1 1

2

1

1

2

β

ata S

NX M

NM

bta S

N

M

e

XX

XX

e

XX

+

+

+

+

/

()

()

/

2

2

2

Multiple Regression 79