In: Economics
In the simple linear regression model, y=Bo +B1x + u, suppose
that E(u) is not equal to 0. Letting a0 = E(u), show
that the model can always be written with the same slope, but a new
intercept and error has a zero expected value.
y = B0 + B1*x + u
E(u) = a0
u = a0 + h, where E(h) = 0
y = B0 + B1*x + a0 + h
y = (B0 + a0) + B1*x + h
here, the new intercept is: (B0 + a0)
The slope is the same, B1
The new error term is h, which has a zero expected value.