Global Portfolio Optimization for BSE Sensex using the Enhanced Black-Litterman Model

摘要

The Markowitz mean-variance optimization algorithm, in conjunction with the enhanced Black Litterman model for estimating expected return of asset returns of Bombay Stock Exchange (BSE), is developed to solve the asset allocation problem. The estimation of expected rate of returns of assets is done by combining economical analysis and technical analysis. The former is done by economists to predict the rate of return based on the present growth of the company and various economic factors while the latter uses past historical data to predict the rate of return. This paper deals with the issues in the prediction of expected rate of return by using the Black Litterman Model which combines both public and private views. The problems of the original Black Litterman Model are analyzed, and the Black Litterman model is enhanced by including the error estimates resulting from the bootstrapping methods. The resulting predicted expected rate of the return vector is given as the input to the Markowitz Mean variance portfolio optimizer to get the better asset allocation model. Bombay Stock Exchange (BSE Sensex) dataset is used and the algorithm is implemented using MATLAB.