Thanks, Glen!

Does anyone know to perform the following algebra: E(B0 + B1*ln(X1) + B2*X2 + e)^2, where "e" is the regression model error term and E is the expectation? I'm trying to replicate whuber's answer here but I do not know how to do the algebra. Just a pointer (a site or material which explain how to do it) would be very helpful. Tks

I think I got it: (B0 + B1ln(X))^2 + 2(B0 + B1ln(X))(B2X2) + (B2X2)^2 + E(e^2). If someone sees a flaw on this, please let me know. Tks