This report discusses a multi-stage stochastic programming model that maximizes expected ending time profit assuming investors can forecast a bull or bear market trend. If an investor can always predict the market trend correctly and pick the optimal stochastic strategy that matches the real market trend, intuitively his return will beat the market performance. For investors with different levels of prediction accuracy, our analytical results support their decision of selecting the highest return strategy. Real stock prices of 154 stocks on 73 trading days are collected. The computational results verify that accurate prediction helps to exceed market return while portfolio profit drops if investors partially predict or forecast incorrectly part of the time. A sensitivity analysis shows how risk control requirements affect the investor's decision on selecting stochastic strategies under the same prediction accuracy. / text

While quantitative investment management has been extensively investigated and many models built in order to provide investment suggestions through quantitative analysis, the combination of quantitative and qualitative analysis is relatively unexplored. The objective of this study is to construct a quantitative stock selection model based on the standard model built by Hsu et al. (2011) which could improve the stability of descriptor and factor structures and the combinability of quantitative and qualitative analysis. The research focuses on the structure of effective factors and descriptors when faced with different types of market trends. Furthermore, we test the performance of a Market Trend-Based Alpha Model (MTB alpha model) and compare with the standard alpha model. The strategy of portfolio construction is a TAIEX enhanced index fund. We find the enhanced index portfolio constructed by the MTB model produces an information ratio of 0.72, which is much higher than the standard model ratio of 0.41. This finding suggests that a MTB model could not only improve performance but also make the descriptor and factor structures more stable and much more easily for managers¡¦ adjusting.

The objective of this study is to build the platform, and the user can choose one model of the three models (1) base multi-factor alpha model (2) sector-specific alpha model (3) market trend-based multi-factor alpha model. The user can choose one target index of the four indexes (1) Electronic (2) Finance (3) Non-Finance Non-Electronics (4) TAIEX. The platform also combined the score of sector-specific model and market model, which we called hybrid model. The platform provides (1) elasticity of equity management (2) completeness of investment strategy (3) inclusiveness of alpha models and target indexes. The user can select a suit model and allocate the model and the target index, and quickly back-testing and evaluate performance. The contributions of this study are that help asset management companies quickly design investment strategies and back-testing, or product many different equity portfolio funds, and evaluate performance for stabled performance.