Simultaneous Equations in Ordered Discrete Responses with Regressor-Dependent Thresholds
The parameters of ordered discrete response (ODR) models are identified only up to a positive scale. In this paper, we examine the identification issue for simultaneous equations with ODR, where the well-known identification problem in simultaneous equations of recovering structural-form parameters from reduced-form parameters is compounded with the ODR identification problem. We allow the thresholds in ODR to be regressor dependent as well as constant; the former is particularly challenging because threshold parameters get mixed with regression parameters, adding one more dimension to the identification problem. We also explore a cross-equation restriction on threshold differences, under which the structural form parameters are fully identified as if the dependent variables are continuously distributed. An empirical example with farm-household joint labour supply is provided to illustrate the identification issues, to show how our proposals work and to apply tests devised for the threshold constancy and cross-equation restrictions. [Econometrics, Mathematical models, Studies
Identification, Ordered discrete response, Simultaneous equations.
Lee, Myoung-jae and Kimhi, A..
Simultaneous Equations in Ordered Discrete Responses with Regressor-Dependent Thresholds. (2005). Econometrics Journal. 8, (2), 176-196. Research Collection School Of Economics.
Available at: http://ink.library.smu.edu.sg/soe_research/500
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
This document is currently not available here.