These questions are at the heart of “Stochastic Frontier Meets Breakdown Frontier,” a research paper by Francisco Rosas, Ph.D. in Economics, conducted in collaboration with Santiago Acerenza and presented at the Annual Economics Conference of the Central Bank of Uruguay (BCU).
This study proposes a methodology for analyzing the robustness of efficiency and productivity estimates when the statistical assumptions used to obtain them are relaxed. It was presented by Rosas at theAnnual Economics Conference of the Central Bank of Uruguay.
Measuring efficiency and testing the results
Stochastic frontier models are one of the tools used to estimate efficiency and productivity. They allow these estimates to be made by taking into account factors such as available technology and inputs used, but to do so, they rely on certain statistical assumptions. The research by Rosas and Acerenza focuses precisely on those assumptions and on what happens to the results when they are relaxed. The goal is to determine to what extent these assumptions can be relaxed without the estimated measures of efficiency ceasing to be valid.
Our approach proposes the extent to which these assumptions can be relaxed or made less strict, so that the estimated efficiency gains remain valid.
Francisco Rosas, Ph.D.
Thus, this study provides a tool for evaluating the robustness of the conclusions drawn from these types of models: it is not only important to have an efficiency estimate, but also to understand the extent to which that estimate holds up when the assumptions on which it was based change.
https://youtu.be/zARpywrGx58