Lesson plan
Generated regressor
In least squares estimation problems, sometimes one or more regressors specified in the model are not observable. One way to circumvent this issue is to estimate or generate regressors from observable data. This generated regressor method is also applicable to unobserved instrumental variables. Under some regularity conditions, consistency and asymptotic normality of least squares estimator is preserved, but asymptotic variance has a different form in general. Suppose the model of interest is the following: : y_{i}=g(x_{1i},x_{2i},\beta)+u_{i} where g is a conditional mean function and its form is known up to finite-dimensional parameter β. Here x_{2i} is not observable, but we know that x_{2i}=h(w_{i},\gamma) for some function h known up to parameter \gamma , and a random sample y_{i}=g(x_{1i},x_{2i},\beta)+u_{i} is available. Suppose we have a consistent estimator \hat\gamma of \gamma that uses the observation w_{i} 's. Then, β can be estimated by (Non-Linear) Least Squares using \hat{x_{2i}}=h(w_{i},\hat\gamma) . Some examples of the above setup include Anderson et al. (1976 and Barro (1977). This problem falls into the framework of two-step M-estimator and thus consistency and asymptotic normality of the estimator can be verified using the general theory of two-step M-estimator.…
Article summary
In least squares estimation problems, sometimes one or more regressors specified in the model are not observable. One way to circumvent this issue is to estimate or generate regressors from observable data. This generated regressor method is also applicable to unobserved instrumental variables. Under some regularity conditions, consistency and asymptotic normality of least squares estimator is preserved, but asymptotic variance has a different form in general. Suppose the model of interest is the following: : y_{i}=g(x_{1i},x_{2i},\beta)+u_{i} where g is a conditional mean function and its form is known up to finite-dimensional parameter β. Here x_{2i} is not observable, but we know that x_{2i}=h(w_{i},\gamma) for some function h known up to parameter \gamma , and a random sample y_{i}=g(x_{1i},x_{2i},\beta)+u_{i} is available. Suppose we have a consistent estimator \hat\gamma of \gamma that uses the observation w_{i} 's. Then, β can be estimated by (Non-Linear) Least Squares using \hat{x_{2i}}=h(w_{i},\hat\gamma) . Some examples of the above setup include Anderson et al. (1976 and Barro (1977). This problem falls into the framework of two-step M-estimator and thus consistency and asymptotic normality of the estimator can be verified using the general theory of two-step M-estimator.…
Key claims
No claims recorded.
Discussion questions
- Summarise the article in your own words. Which single claim would you verify first, and why?
- Which claim in this article is most likely to be revised as the sources update? How would you check its current status?
- If two sources disagreed about one of these facts, what would you need in order to trust one over the other?
Bibliography
No sources recorded.
Key claims and bibliography are generated from the same tracked claims and sources as the article page.