A Stochastic Frontier Model with Endogenous Treatment Status and Mediator
Government policies are frequently used throughout the world to promote productivity. While some of the policies are designed to work through technology enhancement, others are meant to exert the in uence through eciency improvement. For a given technology and eciency channel, another issue to consider in designing and evaluating policies is whether a mediator is required/e ective in achieving the nal outcome. To help better understanding and evaluating the policies, we propose a new stochastic frontier model with treatment effect and mediators. The model allows us to decompose the total program (treatment) effect into the technology and efficiency components, and it also allows us to investigate whether the effect takes place directly from the program or indirectly through a mediator. Both of the treatment status and the mediator are allowed to be endogenous in the fully-parametric model. We illustrate the empirical application of the model using the data of India to study the e ects of large dams on the local districts' agricultural production.
|Date of creation:||Apr 2014|
|Date of revision:||Dec 2014|
|Contact details of provider:|| Phone: 886-2-27822791|
Web page: http://www.econ.sinica.edu.tw/index.php?foreLang=en
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:sin:wpaper:14-a006. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (HsiaoyunLiu)
If references are entirely missing, you can add them using this form.