In this paper, we propose an automatic trading system for portfolio selection that incorporates an investor's trading strategy (aggressive, conservative, or neutral). The system employs technical indicators to forecast assets' future price behaviour. In particular, it clusters assets into three groups: the promising assets are clustered in the “Buy” group, the assets in danger of imminent losses are clustered in the “Sell” group, and the remaining assets are clustered in the “Hold” group. We develop a gradient-based fuzzy rule system that can identify the three groups based on the technical indicator values of the cluster centers. We also develop a labelling algorithm as a corrective measure in case the fuzzy rule-based system identifies more than one group as buy, sell, or hold. Subsequently, we input the clusters to a credibilistic portfolio optimization model that models asset returns using coherent fuzzy numbers. We employ a genetic algorithm to solve the optimization model that exploits the problem's special structure. The proposed methodology is illustrated with a case study of the components of the NASDAQ-100 index

An accurate prediction of asset prices is perhaps the biggest challenge of any study in portfolio optimization. Asset prices are affected by several random and nonrandom factors, which makes them difficult to forecast. This paper proposes a two-phase dynamic portfolio optimization approach. In the first phase, assets are clustered into buy, sell, and hold groups using technical indicators. We provide a methodology to integrate the investor attitude (optimistic, pessimistic, or neutral) during the clustering phase. In the second phase, we input the clustered groups into a portfolio optimization model to obtain the optimum asset allocations. We use coherent fuzzy numbers to model the asset returns to integrate the investor attitude in this phase. The optimization model is solved using a genetic algorithm. The portfolios are rebalanced at regular intervals as new data becomes available. We illustrate the proposed methodology on a 100-asset problem of the US stock market. We analyze the real-world performance of the obtained portfolios. We compare the performance of the proposed approach with the mean–variance model, and other portfolios, such as the naïve portfolio and the NASDAQ-100 index.

In this paper, we propose a new credibility function for a fuzzy variable that can accommodate the attitude of the investor (pessimistic, optimistic, or neutral) along with capturing the return expectations. We use an adaptive index, which the investors can use to specify their general perception of the financial market. We extend the classic mean-variance model so that it provides greater flexibility to the investors in specifying their requirements viz., level of diversification, minimum and maximum level of investment in a particular asset, and the skewness requirement. We also replace variance with mean-absolute semideviation as a measure of quantifying risk, which is more realistic, and solve the resultant multiobjective credibility model with a real-coded genetic algorithm. Numerical examples have been provided at the end to illustrate the methodology and advantages of the model.