I have a model $$Y=\beta_0 + \beta_1 x_1 + \beta_2x_2 +\epsilon$$

I would like the minimum variance unbiased estimate of $\gamma=\beta_1 + \beta_2$. Assuming the Gauss Markov conditions hold, but $x_1$ and $x_2$ are correlated, is there a more efficient way to estimate $\gamma$ than running OLS and adding the estimates of $\beta_1$ and $\beta_2$?

Given that var($\hat{\theta}$)=var($\hat{\beta_1}$)+var($\hat{\beta_2}$)+2cov($\hat{\beta_1},\hat{\beta_2}$) I'm specifically wondering if there is a GLS estimator that reduces cov($\hat{\beta_1},\hat{\beta_2}$) faster than it increases the variance of the individual estimates.


2 Answers 2


The Gauss-Markov theorem states that the covariance matrix of any unbiased estimator $\tilde{\beta} \ne \hat{\beta}_{OLS}$ exceeds that of $\hat{\beta}_{OLS}$ by a positive semidefinite matrix. Let's label the OLS covariance matrix $\Omega$ and the positive semidefinite matrix $D$. The variance of the sum of the OLS estimates can be written as $\iota' \Omega \iota$, where $\iota$ is a vector of ones of the appropriate length. For a non-OLS estimator, the variance of the sum is:

$\sigma^2_\Sigma = \iota' (\Omega + D) \iota = \iota' \Omega \iota + \iota' D \iota \geq \iota' \Omega \iota$

as $D$ is positive semidefinite. Therefore, the sum of the OLS parameter estimates is the minimum variance unbiased estimator of the true sum of the parameters.

In fact, this applies to any weighted sum of the parameter estimates, not just the unweighted sum.


The short answer is no. By the Gauss-Markov theorem, if you want an unbiased estimator, then the OLS is the minimum variance estimator.

However, if you relax the unbiasedness condition, you can get better expected mean square error by using some kind of regularization (e.g. ridge regression).

  • $\begingroup$ The G-M theorem says OLS is the MVUE for the individual parameters. It does not specifically say OLS is MVUE for an estimate of the sum of parameters. $\endgroup$
    – DanB
    Commented Jun 1, 2011 at 21:01
  • $\begingroup$ I think it's valid for any linear combination of the parameters: planetmath.org/encyclopedia/GaussMarkovLinearModel.html $\endgroup$
    – djma
    Commented Jun 2, 2011 at 1:50
  • $\begingroup$ your link does not prove that. $\endgroup$
    – mpiktas
    Commented Jun 2, 2011 at 6:05
  • $\begingroup$ Adding to mpiktas comment, that link says that OLS is a lower variance estimator of y than any other linear combination of variables. It does not discuss the estimation of a linear combination of parameters. $\endgroup$
    – DanB
    Commented Jun 2, 2011 at 17:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.