Solving Triangular Seemingly Unrelated Regression Equations Models on Massively Parallel Systems

Abstract

Solution of triangular seemingly unrelated regression equations (tSURE) models has been investigated. An efficient parallel iterative algorithm based on orthogonal transformations, is proposed for solving an alternative formulation of tSURE models, where the disturbance covariance matrix may be singular. A method to overcome inconsistencies occurring in the tSURE model is also discussed. The major computations of the tSURE algorithm were implemented on a massively parallel computer. A new approach to construct accurate execution time models is used for analysing the performance of parallel implementations.KeywordsExecution TimeAmerican Statistical AssociationParallel SolutionTriangular SystemGood Linear Unbiased EstimatorThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.