Abstract
Productivity analysis plays a pivotal role in evaluating the performance of various economic entities, especially in the context of sustainable development. Since productivity measures are based on the notion of output-input ratios, non-positive values of inputs and outputs are not permissible. Overtime, various approaches to the imposition of the strict positivity assumption have emerged. However, there are no clear guidelines on the choice of different methods of handling this econometric problem under various circumstances. This paper provides an empirical review of the strict positivity assumption as it applies to productivity analysis, specifically focusing on its application to stochastic frontier framework. We examine the existing lines of approaches to handling zero values of inputs and outputs. Additionally, we empirically present a comprehensive case study using a real-world dataset together with Monte Carlo simulations to demonstrate the implications of various approaches in eco-efficiency stochastic frontier models. Finally, on the basis of our findings, we provide recommendations for future research and practical applications.
About the author:
Sulman Owili
Sulman is a Masters student at the Department of Agricultural Economics, University of Nairobi, Kenya. He is also a research fellow at the International Institute of Tropical Agriculture (IITA). In August 2023, he was awarded the Best CMAAE Student in the 2022 SFSE. Sulman holds a bachelors degree in Agricultural Education and Extension from the same department.