If residuals for the same observation from two 'seemingly unrelated' equations are correlated, then it is often stated that there are efficiency gains from estimating the parameters of the two equations jointly using Zellner’s (1962) over separate OLS.

Gains on 'efficiency' grounds is the reasoning presented in this IMF paper for example examples, for using SUR over OLS on each estimating equation.

However, the intuition behind the so-called improved efficiency of SUR over OLS, when cross-equation contemporaneous residuals are correlated, is not clear to me.

Something I thought is that because SUR system, by construction, combines information on different equations it utilizes more observations at once than OLS does and more observations are generally welcomed. Still, the idea that SUR improves OLS on efficiency grounds is not clear to me. Is it that parameters estimated from SUR will have have smaller standard errors than those from OLS? Will the SUR equations have smaller RMSE than the OLS equations?

Any insights on how to think about this would be helpful.

  • 2
    $\begingroup$ Do you understand why generalized least squares (GLS) is more efficient than OLS when the errors have nonzero correlations? SUR is just a Feasible GLS with a special structure imposed on the error covariance matrix. $\endgroup$
    – jbowman
    Feb 10, 2020 at 17:44

1 Answer 1


Your fallacy is thinking that "efficiency" merely boils down to number of observations. SUR does not "utilize more observations" than OLS. The relative efficiency of two estimators is the ratio of their squared standard errors. You must also consider the bias of the estimators and distinguish between attributes of regressors ($\beta$) and the resulting predictions. The RMSE is generally discussed as a measure of predictive accuracy, e.g. the variance of the residuals.

When the linear model is correctly specified, any arbitrarily weighted linear regression model, including the "unweighted" OLS, is unbiased. This holds even when the "independence of errors" assumption is violated. However, the Gauss-Markov theorem tells us the only BLUE (Best Linear Unbiased Estimator) is the inverse-variance weighted linear regression.

SUR (seemingly unrelated regressions) proposes to use the (inverse of the) first-step estimate of the covariance structure as a plug-in weight for the second-step regression. Provided the appropriate assumptions hold, this will do us better than pretending there is no between-regression variance as would be the case by fitting a bunch of OLS models and saying, "done". However, both of these are still less efficient than simply knowing the between regression variance and fixing the weights accordingly. This is the only model that would, in finite samples (and asymptotically?), achieve the Cramer-Rao lower bound.

  • $\begingroup$ Thank you. Two follow-up questions: 1) From what you explained above, is it then the case that the SE of an OLS coefficient for variable x will always be lower than than SE of an SUR coefficient for the same variable, whenever the equations in the system are related? 2) SUR boils down to FGLS? $\endgroup$ Feb 18, 2020 at 1:36
  • 1
    $\begingroup$ @StatsScared no, not always lower. "Achieving the Cramer Bound" is a property of long-run expectation, so for some samples it may be the case that comparing weighted and unweighted, the unweighted model "looks" better: just an artifact of design. Can't comment much on 2, but it seems plausible. $\endgroup$
    – AdamO
    Feb 18, 2020 at 16:00
  • $\begingroup$ great point. Interestingly enough, a rarely made one. Even Stata's documentation for SUR estimation fails to make the distinction you just made, which without further knowledge on the topic can be confusing. $\endgroup$ Feb 18, 2020 at 21:30

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.