Current browse context:
q-fin.PM
Change to browse by:
References & Citations
Quantitative Finance > Portfolio Management
Title: Predictive Decision Synthesis for Portfolios: Betting on Better Models
(Submitted on 30 Apr 2024)
Abstract: We discuss and develop Bayesian dynamic modelling and predictive decision synthesis for portfolio analysis. The context involves model uncertainty with a set of candidate models for financial time series with main foci in sequential learning, forecasting, and recursive decisions for portfolio reinvestments. The foundational perspective of Bayesian predictive decision synthesis (BPDS) defines novel, operational analysis and resulting predictive and decision outcomes. A detailed case study of BPDS in financial forecasting of international exchange rate time series and portfolio rebalancing, with resulting BPDS-based decision outcomes compared to traditional Bayesian analysis, exemplifies and highlights the practical advances achievable under the expanded, subjective Bayesian approach that BPDS defines.
Link back to: arXiv, form interface, contact.