Inconsistencies in SURE Models: Computational Aspects

Authors:
Erricos J. Kontoghiorghes
Affiliations:
Institut d'informatique, Université/ de Neuchâ/tel, Rue Emile-Argand 11, CH-2007 Neuchâ/tel, Switzerland/ E-mail: erricos.kontoghiorghes@info.unine.ch
Venue:
Computational Economics - Special issue on computational studies at Cambridge
Year:
2000

Citing 4
Cited 0

Seemingly unrelated regression equations models

Seemingly unrelated regression equations models
An alternative approach for the numerical solution of seemingly unrelated regression equations models

Computational Statistics & Data Analysis
Computing 3SLS Solutions of Simultaneous Equation Models with a Possible Singular Variance–Convariance Matrix

Computational Economics - Special issue on computational economics in Geneva: volume 1: computational econometrics, statistics, and optimization
Parallel Strategies for Solving SURE Models with Variance Inequalities and Positivity of Correlations Constraints

Computational Economics - Computational Studies at Stanford

Quantified Score

Hi-index	0.00

Visualization

Abstract

The solution of the SURE model with singular variance-covariance matrixresults in redundancies and possibly inconsistencies among the observationsof the model. A numerical procedure is proposed and investigated thatgenerates a consistent model from an inconsistent one. The use of SVD has beenused to compute the various factorizations arising in the solution of the SUREmodel when treated as a generalized linear least squares problem.