[4Xin2-55] SBLM: Sparse Black Litterman Model using the Spike and Slab Prior
Keywords:Portfolio Theory, Black Litterman Model, Sparse Modelilng
The Black-Litterman (BL) method is a useful tool for portfolio construction, treating expected returns and investors' outlook as random variables, and estimating asset weights through Bayesian updating. However, the BL method involves numerous parameters that can influence asset weights, potentially leading to excessive rebalancing, increased transaction costs, and deteriorated performance.
Therefore, in this study, we propose the Sparse Black-Litterman (SBL) method to reduce transaction costs.
Specifically, by incorporating the Spike and Slab prior distribution into weight changes, we introduce sparsity to weight fluctuations, thereby suppressing unnecessary rebalancing.
This approach allows for the construction of an efficient portfolio that integrates investor views while reducing transaction costs. Theoretically, we prove that the introduction of a prior distribution can be reduce rebalancing.
In the empirical analysis, we use both synthetic and real data to validate the effectiveness of our proposed method and its impact on reducing transaction costs.
Therefore, in this study, we propose the Sparse Black-Litterman (SBL) method to reduce transaction costs.
Specifically, by incorporating the Spike and Slab prior distribution into weight changes, we introduce sparsity to weight fluctuations, thereby suppressing unnecessary rebalancing.
This approach allows for the construction of an efficient portfolio that integrates investor views while reducing transaction costs. Theoretically, we prove that the introduction of a prior distribution can be reduce rebalancing.
In the empirical analysis, we use both synthetic and real data to validate the effectiveness of our proposed method and its impact on reducing transaction costs.
Please log in with your participant account.
» Participant Log In