Journal of Mathematical Sciences: Advances and ApplicationsAuthor's: D. A. Agunbiade
Pages: [115] - [125]
Received Date: June 30, 2017
Submitted by:
DOI: http://dx.doi.org/10.18642/jmsaa_7100121861
The inconsistent estimators of the parameters resulted from the simultaneity problem when the structural equations are estimated by ordinary least squares (OLS) approach is of great importance in this research. The study argues that it does not mean that OLS estimation of simultaneous equation models (SEM) is useless; and so examined the empirical analysis of the three categories of estimation techniques: OLS, single equation and system estimation approaches. The asymptotic and small sample properties of the various methods are compared using total absolute bias and the root mean square error. The findings indicated no remarkable and asymptotic pattern in the performance of the estimates of the parameters of each estimator. However, the ranking of the three categories of simultaneous estimation methods differs according to the correlation level of the explanatory variables and the individual equation of the SEM.
Monte Carlo methods, simultaneous equation models, estimating techniques, regression models, OLS approach.
