Consequences of Ignoring Dependent Error Components and Heterogeneity in a Stochastic Frontier Model: An Application to Rice Producers in Northern Thailand

The traditional Stochastic Frontier Model (SFM) suffers from a very restrictive assumption of independence its error components and also limited ability to address heterogeneity (inefficiency effects) satisfactorily, thereby leading potential biases in the estimation model parameters, identification inefficiency effect variables influencing efficiency and, ultimately, scores. This paper aims investigate consequences ignoring any dependency stochastic frontier model, proposes copula-based SFM with resolve such weaknesses based on simulation study prove superiority over SFM, followed by an empirical application sample rice producers northern Thailand. We demonstrate that proposed i.e., dependent heterogeneity, is unbiased robust. experiments show can cause parameter severe overestimation technical efficiency. has similar consequences. However, just does not have great impact compared consequence components. results land, labor material inputs are all significant drivers production our whereas only land variable seems be driver production. mean (MTE) score was overestimated two points MTE = 0.88 versus 0.86. Finally, reveals both subsistence pressure use hired significantly associated inefficiency, could identify only. Therefore, caution necessary when interpreting conventional as may biased, incomplete and/or inadequate.