# Why can we assume that the expected value of the error term is zero? [closed]

Why can we assume that the expected value of the error term in a linear regression model is zero?

This is with regard to a simple linear regression.

-

## closed as unclear what you're asking by Nick Peterson, vonbrand, T. Bongers, egreg, mathematics2x2lifeFeb 11 '14 at 23:39

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question.If this question can be reworded to fit the rules in the help center, please edit the question.

Don't know really what you mean, but if the expected error were nonzero, adjusting the regression function accordingly to lower the expectation would mean an improvement of the matching the regression does – Hagen von Eitzen Feb 11 '14 at 21:30

Because you estimate a constant term $\alpha$ in simple linear regression $$y_i=\alpha+\beta x_i+\varepsilon_i,$$ so that $E[\varepsilon_i]\neq 0$ just means $\alpha$ is replaced by $\hat{\alpha}=\alpha-E[\varepsilon_i]$. Hence, you might as well assume the error term has mean zero.