A Monte Carlo Comparison of Asymmetric Price Transmission Linear Models Under the Effect of Difference in Asymmetric Adjustment Parameters using rMDL, gMDL, nMDL, eMDL, AIC & BIC

Abstract

The effect of asymmetric adjustment on the ability of variants of minimum description length criteria was investigated in their ability to recover the true Data Generating Process. A Monte Carlo method was employed within the linear Asymmetric Price Transmission modelling framework. Results indicated empirically that application of a more complex model or increase in the number of asymmetric adjustment parameters improves the recovery of the true data generating process by the model selection methods. Minimum Description Length criteria are very reliable and useful in Asymmetric Price Transmission modelling. Researchers should always aim at stronger levels of asymmetry to achieve optimal Asymmetric Price Transmission linear models.