Page 195 - Microsoft Word - 00 ACCA F2 Prelims.docx

Statistical techniques

Linear regression

With the high-low method only two observations are used and a linear relationship is
assumed. However these two extreme points may hide the true relationship between
the variables.

y
ion

x
Consider the equation of a straight line:

y = a + bx

where: y = dependent variable

x = independent variable

a = intercept on y-axis

b = gradient of the line

189