When making regression predictions for Y observations based on X observations, one uses the regression line for Y on X. While all such estimates can be achieved using only the slope of the regression line
some people prefer to use the equation of the regression line. To obtain this equation, you need to note the slope of the line [as above] and remember that the line passes through the center of averages (AX , AY). Using the most common form of a linear equation Y = (slope) X + (Y-intercept) we immediately see that
To find the Y-intercept, we use the fact that the regression line passes through (AX , AY). This tells us
Solving this expression for the Y-intercept, the equation of the regression line for Y on X becomes
Using some basic algebra, this equation has an equivalent expression:
With a bit more algebra, we see:
Recalling the conversions to standard units for both X and Y observations, we see from this expression that the equation of the regression line for Y on X in terms of standard units is
Last revised: February 2000. Questions to: kvalente@mail.colgate.edu
Copyright 2000 © Colgate University. All
rights reserved.