FINANCE Corporate financial policy and R and D Management

term, ei, is the difference between the actual and estimated dependent vari-

able values for any given independent variable values, Xi.

(5.3)

The regression error term, ei, is the least squares estimate of εi, the actual

error term.

To minimize the error terms, the least squares technique minimizes the

sum of the squares error terms of the Nobservations,

(5.4)

The error terms from the Nobservations will be minimized. Thus, least

squares regression minimizes:

e

i

N

1

2

= 1

∑

eYYiii=−ˆ

74 AN INTRODUCTION TO STATISTICAL ANALYSIS AND SIMULTANEOUS EQUATIONS

FIGURE 5.3 The Regression Model

Population

regression line

e 3 = 1

e 3 = 1^1 / 2

A + BX

A + BX 2

A + BX 1

X 1 X 2

Y 1

Y 1

Y