Bayesian median autoregressive models for consumer price index prediction

Consumer price index is one of the most widely recognized macroeconomic indicators as it is associated with various economic situations.In this study, we aim to predict the consumer price index in Korea using a Bayesian median autoregressive model.Subsequently, we compare the performance of the consumer price index predicted by the Bayesian median autoregressive model with other frequentist and mean-based autoregressive models to determine which model performs the best.Furthermore, to reflect the uncertainty of model selection, we utilize Bayesian model averaging for prediction.The results indicate that the predictive accuracy of the proposed models surpasses that of other comparison models.Therefore, it was confirmed that the Bayesian median autoregressive model can serve as an alternative model to frequentist and mean-based autoregressive models in predicting the consumer price index.