Estimation of SUR Model with Non-Nested Missing Observations Academic Article uri icon

abstract

  • This paper considers alternative two-step estimators and their small sample properties for the seemingly unrelated regression (SUR) model with non-nested missing observations. A Monte Carlo experiment indicates that alternative estimators have more profound differences in their efficiency, compared to the case of nested missing observations. In particular, the two-step application of the Hartley-Hocking maximum likelihood estimator can realize a significant gain in efficiency. There are substantial losses in efficiency when only the subset of data that has complete observations is used in estimation. /// Cet article prend en considration des estimateurs de remplacement ; deux chelons et leurs petites proprits d'chantillon pour le modle "Rgression en apparence sans rapports" (Seeminly Unrelated Regression, SUR) avec des observations manquantes nonimbriques. Une exprience ; la Monte Carlo indique que des estimateurs de remplacement ont des diffrences plus notables quant ; leur efficacit compar aux cas d'observations manquantes imbriques. En particulier, l'application ; deux chelons de l'estimateur ; probabilit maximale Hartley-Hocking peut raliser d'apprciables gains en efficacit. Il n'y a perte considrable de gains d'efficacit uniquement lorsqu'on emploie dans l'estimateur un sous-ensemble de donnes ayant des observations compltes.

published proceedings

  • Annals of Economics and Statistics

author list (cited authors)

  • Hwang, .., & Schulman.

citation count

  • 2

publication date

  • January 1996

publisher