Publication details
The Kitchen Sink Problem: Monte Carlo study of stochastic frontiers
| Authors | |
|---|---|
| Year of publication | 2015 |
| Type | Article in Proceedings |
| Conference | 33rd International Conference Mathematical Methods in Economics Conference Proceedings |
| MU Faculty or unit | |
| Citation | |
| Web | Conference proceedings |
| Field | Economy |
| Keywords | Monte Carlo Simulation; problem of identification; problem of collinearity; stochastic frontier analysis; nested models; model selection |
| Description | This article presents a Monte Carlo study of the problem of identification that arises when superfluous variables are added into the regression that are correlated with other regressors. The results show that the bias in coefficients that occurs is substantial even in large samples and that the standard tests for favourability of the nested models are actually likely to reject the restricted model in cases when the bias is particularly large. It is also noted that in the context of frontier analysis, this problem can be partially mitigated when attention is directed specifically to the characteristics of the production process rather than the coefficients of the frontier themselves. |
| Related projects: |