Generalized least squares

In statistics, generalized least squares (GLS) is a technique for estimating the unknown parameters in a linear regression model when there is a certain degree of correlation between the residuals in a regression model. In these cases, ordinary least squares and weighted least squares can be statistically inefficient, or even give misleading inferences. GLS was first described by Alexander Aitken in 1936.[1]

Method outlineEdit

In standard linear regression models we observe data   on n statistical units. The response values are placed in a vector  , and the predictor values are placed in the design matrix  , where   is a vector of the k predictor variables (including a constant) for the ith unit. The model forces the conditional mean of   given   to be a linear function of  , and assumes the conditional variance of the error term given   is a known nonsingular covariance matrix  . This is usually written as


Here   is a vector of unknown constants (known as “regression coefficients”) that must be estimated from the data.

Suppose   is a candidate estimate for  . Then the residual vector for   will be  . The generalized least squares method estimates   by minimizing the squared Mahalanobis length of this residual vector:


Since the objective is a quadratic form in  , the estimator has an explicit formula:



The GLS estimator is unbiased, consistent, efficient, and asymptotically normal with   and  . GLS is equivalent to applying ordinary least squares to a linearly transformed version of the data. To see this, factor  , for instance using the Cholesky decomposition. Then if we pre-multiply both sides of the equation   by  , we get an equivalent linear model   where  ,  , and  . In this model  , where   is the identity matrix. Thus we can efficiently estimate   by applying OLS to the transformed data, which requires minimizing


This has the effect of standardizing the scale of the errors and “de-correlating” them. Since OLS is applied to data with homoscedastic errors, the Gauss–Markov theorem applies, and therefore the GLS estimate is the best linear unbiased estimator for β.

Weighted least squaresEdit

A special case of GLS called weighted least squares (WLS) occurs when all the off-diagonal entries of Ω are 0. This situation arises when the variances of the observed values are unequal (i.e. heteroscedasticity is present), but where no correlations exist among the observed variances. The weight for unit i is proportional to the reciprocal of the variance of the response for unit i.[2]

Feasible generalized least squaresEdit

If the covariance of the errors   is unknown, one can get a consistent estimate of  , say  ,[3] using an implementable version of GLS known as the feasible generalized least squares (FGLS) estimator. In FGLS, modeling proceeds in two stages: (1) the model is estimated by OLS or another consistent (but inefficient) estimator, and the residuals are used to build a consistent estimator of the errors covariance matrix (to do so, one often needs to examine the model adding additional constraints, for example if the errors follow a time series process, a statistician generally needs some theoretical assumptions on this process to ensure that a consistent estimator is available); and (2) using the consistent estimator of the covariance matrix of the errors, one can implement GLS ideas.

Whereas GLS is more efficient than OLS under heteroscedasticity or autocorrelation, this is not true for FGLS. The feasible estimator is, provided the errors covariance matrix is consistently estimated, asymptotically more efficient, but for a small or medium size sample, it can be actually less efficient than OLS. This is why, some authors prefer to use OLS, and reformulate their inferences by simply considering an alternative estimator for the variance of the estimator robust to heteroscedasticity or serial autocorrelation. But for large samples FGLS is preferred over OLS under heteroskedasticity or serial correlation.[3] [4]A cautionary note is that the FGLS estimator is not always consistent. One case in which FGLS might be inconsistent is if there are individual specific fixed effects.[5]

In general this estimator has different properties than GLS. For large samples (i.e., asymptotically) all properties are (under appropriate conditions) common with respect to GLS, but for finite samples the properties of FGLS estimators are unknown: they vary dramatically with each particular model, and as a general rule their exact distributions cannot be derived analytically. For finite samples, FGLS may be even less efficient than OLS in some cases. Thus, while GLS can be made feasible, it is not always wise to apply this method when the sample is small. A method sometimes used to improve the accuracy of the estimators in finite samples is to iterate, i.e. taking the residuals from FGLS to update the errors covariance estimator, and then updating the FGLS estimation, applying the same idea iteratively until the estimators vary less than some tolerance. But this method does not necessarily improve the efficiency of the estimator very much if the original sample was small. A reasonable option when samples are not too large is to apply OLS, but throwing away the classical variance estimator


(which is inconsistent in this framework) and using a HAC (Heteroskedasticity and Autocorrelation Consistent) estimator. For example, in autocorrelation context we can use the Bartlett estimator (often known as Newey-West estimator since these authors popularized the use of this estimator among econometricians in their 1987 Econometrica article), and in heteroskedastic context we can use the Eicker–White estimator. This approach is much safer, and it is the appropriate path to take unless the sample is large, and "large" is sometimes a slippery issue (e.g. if the errors distribution is asymmetric the required sample would be much larger).

The ordinary least squares (OLS) estimator is calculated as usual by


and estimates of the residuals   are constructed.

For simplicity consider the model for heteroskedastic errors. Assume that the variance-covariance matrix   of the error vector is diagonal, or equivalently that errors from distinct observations are uncorrelated. Then each diagonal entry may be estimated by the fitted residuals   so   may be constructed by


It is important to notice that the squared residuals cannot be used in the previous expression; we need an estimator of the errors variances. To do so, we can use a parametric heteroskedasticity model, or a nonparametric estimator. Once this step is fulfilled, we can proceed:

Estimate   using   using[4] weighted least squares


The procedure can be iterated. The first iteration is given by


This estimation of   can be iterated to convergence.

Under regularity conditions any of the FGLS estimator (or that of any of its iterations, if we iterate a finite number of times) is asymptotically distributed as


where n is the sample size and


here p-lim means limit in probability

See alsoEdit


  1. ^ Aitken, A. C. (1936). "On Least-squares and Linear Combinations of Observations". Proceedings of the Royal Society of Edinburgh. 55: 42–48.
  2. ^ Strutz, T. (2016). Data Fitting and Uncertainty (A practical introduction to weighted least squares and beyond). Springer Vieweg. ISBN 978-3-658-11455-8., chapter 3
  3. ^ a b Baltagi, B. H. (2008). Econometrics (4th ed.). New York: Springer.
  4. ^ a b Greene, W. H. (2003). Econometric Analysis (5th ed.). Upper Saddle River, NJ: Prentice Hall.
  5. ^ Hansen, Christian B. (2007). "Generalized Least Squares Inference in Panel and Multilevel Models with Serial Correlation and Fixed Effects". Journal of Econometrics. 140 (2): 670–694. doi:10.1016/j.jeconom.2006.07.011.

Further readingEdit