1.Here is an example of “least squares” regression being used to predict total cholesterol level (“y”) from body mass index (“x”).

2.In this example, we provide the correlation coefficient (r) between total cholesterol level and body mass index in order to calculate the b1 term.

3.Notice the equation of the regression line. The constant, b0, is the theoretical value of total cholesterol when the “x” variable, body mass index, is zero.

4.However, since body mass index cannot be zero, the b0 term is purely theoretical and is used as constant to generate predicted values of total cholesterol based on body mass index.