|
|
|
|
|
Instrumental VariablesThe technique of instrumental variables is used in regression analysis. A standard assumption of regression is that the explanatory variables are uncorrelated with the unexplained component. When that assumption fails, ordinary linear regression will not provide consistent estimates. However, if an instrument is available, consistent estimates may still be obtained. An instrument is a variable that does not itself belong in the regression, that is correlated with the explanatory variable, and that is uncorrelated with the error term. For the regression equation the standard assumption is that The method of moments estimating equation would then be This produces the estimator When x and are uncorrelated, the final term vanishes in the limit providing a consistent estimator. When the standard assumption fails, an instrument z known to be uncorrelated with the error term can be used. This provides the instrumental variables moment condition which produces consistent estimates. This produces the estimator When z and are uncorrelated, the final term vanishes in the limit providing a consistent estimator. Classic examples of the technique include the errors in variables problem, and the recovery of structural parameters from simulataneous equations models such as supply and demand.
|
 |
|
| Copyright 2005-2009 OnPedia.com. All Rights Reserved |
|
|