Files
Abstract
Akaike’s Information Criteria provide a basis for choosing between competing approaches to testing for
price asymmetry. However, very little research has been undertaken to understand its performance in
the price transmission modelling context. In addressing this issue, this paper introduces and applies parametric
bootstrap techniques to evaluate the ability of Akaike Information Criteria (AIC) and Consistent Akaike
Information Criteria (CAIC) in distinguishing between competing asymmetric price transmission models
under various error and sample size conditions. Bootstrap simulation results suggest that the performance
of the model selection methods depends on sample size and stochastic variance. The Bootstrap simulations
further indicate that CAIC is consistent and performs better than the AIC in large bootstrap samples.
The ability of the model selection methods to identify the true asymmetric price relationship decreases with
increase in stochastic variance. The research findings demonstrate the usefulness of Bootstrap algorithms
in price transmission model comparison and selection.