Portfolio Optimization Based on an Improved Black-Litterman Model
For the"imperfect effectiveness"of financial markets and the"imperfect rationality"of inves-tors,an improved Black-Litterman model combining investor views and multi-channel information is de-veloped by means of the Bayesian framework to determine the optimal personalized investment strate-gies.In an empirical study of the Chinese stock market,the problem of quantifying investor views is ad-dressed using the SVM-AR1MA-GARCH model.Compared with several reference strategies,the optimal investment strategy identified by the improved Black-Litterman model has a more robust out-of-sample performance,with higher Sharpe ratios and lower turnover ratios in different market conditions.