A hybrid portfolio theory model based on genetic algorithm and vector quantization

Abstract

Portfolio theory considers a sequential portfolio selection procedure with the goal of best analyzing and predicting the performance of the market. Data compression techniques such as context-based modeling and vector quantization are based on a memoryless or near-memoryless prediction process and suit for the memoryless nature of the stock market data. In this paper, we propose a data compression-based portfolio prediction model hybridized with the fuzzy logic and genetic algorithm. This model exploits the prediction power of the lossy compression techniques such as vector quantization and context-based modeling and applies the prediction to the portfolio theory. In the model, the quantifiable microeconomic stock data are first optimized through the genetic algorithms to generate the most effective microeconomic data in relation to the stock market performance. The nonquantifiable microeconomic data are modeled with the fuzzification process. Then the previous stock market performance with the effective stock data and the fuzzified microeconomic data are processed based on the context-based modeling and vector quantization. Finally, the prediction of the stock market performance with the stock data is defuzzified using the fuzzification model to produce a portfolio performance prediction. The experiment on JF Asean Unit Trust using this portfolio theory model has shown a reliable prediction for the performance of the Trust for the past 5 years.