NONLINEAR ESTIMATION OF THE CES PRODUCTION FUNCTION: SAMPLING DISTRIBUTIONS AND TESTS IN SMALL SAMPLES.

In this paper the results of the Monte Carlo experiments presented in [8] are resurrected to estimate the probability density functions of the nonlinear least-squares estimators of the CES parameters. Further, the earlier results are used to estimate the densities of the Student-t and Durbin-Watson...

Descripción completa

Detalles Bibliográficos
Publicado en:Southern Economic Journal Vol. 41; no. 2; pp. 258 - 267
Autores principales: Kumar, T. Krishna, Gapinski, James H.
Formato: Artículo
Publicado: Wiley-Blackwell Oct74
Materias:
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:In this paper the results of the Monte Carlo experiments presented in [8] are resurrected to estimate the probability density functions of the nonlinear least-squares estimators of the CES parameters. Further, the earlier results are used to estimate the densities of the Student-t and Durbin-Watson statistics, the values of which emerged as by-products of the regression process. The latter densities permit us to shed some light on the appropriateness of the corresponding tests in the nonlinear context. This paper therefore stands as a material extension of [8]. <BR> This study investigated the nature of small sample density functions of nonlinear least-squares regression estimators of the CES production parameters by means of the Monte Carlo technique. The information on such density functions prompted an inquiry into the appropriateness of the Student-t test of significance originally intended for least-squares estimators in linear regression models. An approximation to the sampling distribution of the Durbin-Watson statistic was obtained for the nonlinear regression model.