Biometrical Letters Vol. 44(2), 2007, pp. 97-103
|
NON-OBSERVABLE REGRESSORS
J.T. Mexia1 and M.M. Oliveira2 1Department of Mathematics, Faculty of Science and Technology. New University of Lisbon. Monte da Caparica 2829-516 Caparica Portugal. 2Department of Mathematics. University of Évora. Colégio Luis António Verney,Rua Romao Ramalho 59, 7002. Évora. Portugal. Email: mmo@uevora.pt |
An example is used to show how, in clearly formulated problems, non-observable regressors may emerge. The model matrices and goal functions for the adjustment are discussed. The propose being the adjustment of not only the regression coefficients but also the non-observable regressors. Algorithms are presented to carry out the adjustments. It is discussed how the adjustment may be validated through analysis of the residues.
Non-observable regressor, regression coefficients, Algorithms.
