What is it about?
This paper provides one-stop solutions for economic research using instrumental variables. Using our STATA user defined command naivereg, empirical researchers no longer need to worry about which instrument is strong or weak, and our method is general enough to take care of different functional form of the reduced form equations. Our method is as good as ivreg (ivreg2) when the functional form is linear, and much better than those when the functional form is nonlinear.
Featured Image
Why is it important?
In empirical studies, instrumental variable method is commonly used to solve endogeneity problem. But when facing real economic problems, we often do not know 1, which instrument to include in the model; 2, which functional form do we assume. Our proposed estimator addresses the two concerns simultaneously, by improving the finite sample performance of IV estimator and is most efficient in large samples.
Perspectives
This paper addresses the instrumental variable selection and estimation issues for empirical researchers. With a very convenient computation algorithm and better finite sample performance against the conventional linear IV estimator, I believe this new estimator should be among the regular toolbox for empirical studies employing instrumental variables in the future.
Dr. Qingliang Fan
Chinese University of Hong Kong
Read the Original
This page is a summary of: Nonparametric Additive Instrumental Variable Estimator: A Group Shrinkage Estimation Perspective, Journal of Business and Economic Statistics, April 2017, Taylor & Francis,
DOI: 10.1080/07350015.2016.1180991.
You can read the full text:
Contributors
The following have contributed to this page
