Treatment of Endogeneity in the estimation of a production function with time-varying parameters

Tran, Yen Thi Hai (2014). Treatment of Endogeneity in the estimation of a production function with time-varying parameters Honours Thesis, School of Economics, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Yen_Thi_Hai_Tran_Honours.pdf Full text application/pdf 1003.74KB 11
Author Tran, Yen Thi Hai
Thesis Title Treatment of Endogeneity in the estimation of a production function with time-varying parameters
School, Centre or Institute School of Economics
Institution The University of Queensland
Publication date 2014-11-11
Thesis type Honours Thesis
Supervisor Alicia Rambaldi
Antonio Peyrache
Total pages 98
Language eng
Subjects 14 Economics
Formatted abstract
Endogeneity arises in the estimation of production functions due to a number of reasons. Simultaneity endogeneity due to the joint determination of the input levels and the unobserved productivity level has been addressed in the literature. However, endogeneity can also arise from measurement errors and the correlation between the level of inputs and their time-varying marginal productivity. These latter sources of endogeneity have not received much attention to date. This thesis proposes an unobserved components model that can account for all three types of endogeneity and the time-variation in the structural parameters of the production function. The standard Kalman filter coupled with maximum likelihood can consistently estimate the structural parameters when the marginal productivity of the inputs are constant. In the time-varying marginal productivity case, we propose a two-step filter that involves the Dynamic Conditional Score filter and the standard Kalman filter. The form of the estimators is derived to show they can estimate the structural parameters without making restrictive assumptions or employing computationally intensive methods. The estimators' performance is assessed in a simulation study that covers all endogeneity scenarios. The results show the proposed estimators can recover the true parameters in the presence of severe endogeneity conditions and capture a large range of parameter dynamics.
Keyword Endogeneity
Production function

Document type: Thesis
Collection: UQ Theses (non-RHD) - UQ staff and students only
 
Citation counts: Google Scholar Search Google Scholar
Created: Mon, 06 Jul 2015, 13:34:19 EST by Mr Yun Xiao on behalf of Scholarly Communication and Digitisation Service