Nonlinear Vector Error Correction Models in Price Transmission Analysis: Threshold Models vs. Markov-Switching Models

This work provides a comparison of methodologies for applied research in price transmission analysis. We compare two regime-dependent econometric models, namely the threshold vector error correction model and the Markov-switching vector error correction model. We first provide a conceptual comparison in which we find that the regime-switching mechanisms of the models differ fundamentally so that each model is suitable for a certain type of nonlinear price trans-mission. Furthermore, we conduct a Monte Carlo experiment in order to study the performance of each of the models’ estimation techniques for simulated data. Although each model possesses an immediate economic interpretation which well matches an aspect of the theory of price transmission, the simulation results indicate that the corresponding estimation techniques yield biased estimates of low precision.