Description:
This study introduces a novel Bayesian switching stochastic frontier model for predicting the direction of investment means, such as stocks, cryptocurrencies, and precious metals. The model considers three market states: bear, consolidation, and bull. We demonstrate the model's applicability using Bitcoin as a case study, analyzing its price behavior based on various blockchain metrics. Our findings uncover key relationships between these metrics and market trends, offering valuable insights for investors. The proposed estimation strategy contributes significantly to the field of econometric modeling and enhances the understanding of market dynamics