Bayesian Model Averaging and Prior Sensitivity in Stochastic Frontier Analysis

Abstract

This paper discusses Bayesian model averaging (BMA) in Stochastic Frontier Analysis and investigates inference sensitivity to prior assumptions made about the scale parameter of (in)efficiency. We turn our attention to the “standard” prior specifications for the popular normal-half-normal and normal-exponential models. To facilitate formal model comparison, we propose a model that nests both sampling models and generalizes the symmetric term of the compound error. Within this setup it is possible to develop coherent priors for model parameters in an explicit way. We analyze sensitivity of different prior specifications on the aforementioned scale parameter with respect to posterior characteristics of technology, stochastic parameters, latent variables and—especially—the models’ posterior probabilities, which are crucial for adequate inference pooling. We find that using incoherent priors on the scale parameter of inefficiency has (i) virtually no impact on the technology parameters; (ii) some impact on inference about the stochastic parameters and latent variables and (iii) substantial impact on marginal data densities, which are crucial in BMA.

A Method to Explore the Best Mixed-Effects Model in a Data-Driven Manner with Multiprocessing: Applications in Public Health Research

“…Although these methods perform effectively in generating prediction models, compared with the model exploration method I will propose, they have several limitations when used in health and psychological research. The result from BMA does not suggest one specific best model; instead, it demonstrates coefficients from averaging multiple candidate models [33]. Furthermore, gathering information for statistical inference, such as significance, by performing regularization is more difficult than conventional analysis methods [34].…”