الفهرس 
Estimation of simultaneous equations models with measurement errors
SummaryOnly 14 pages are availabe for public view
 |  | from | 172 |  |  |
| | | from | 172 | | |
Abstract
Economic variables are often measured with error, if they have been neglected and ignored the measurement errors ME, so the results of the estimation based on traditional methods such as (OLS) be biased and inconsistent, and so we get statistic estimates ineffective and misleading inferences. And a result of this cannot be trusted and depended on making any decision. This study is interested with estimation methods of simultaneous equations models SEM with the presence of measurement errors ME in the exogenous variables in the single structural equation. And we assumed that the ME is random and uncorrelated. As there are two methods to estimate the SEM with the ME, namely: Estimation by full information methods (FIM) and limit information methods (LIM) (i.e.: estimating of each equation separately without looking at the rest of the equations in the system), and the latter is used in this study. We depended on two methods to estimate the parameters of the single structural equation, namely limit information maximum likelihood principle and two stage least squares 2 - SLS