許多學者透過量化多因子模型的方式，企圖以實證的角度去尋找那些績效存在規律，且能夠為投資人帶來超額報酬的因子，而除了單純的因子選股外，也有文獻開始加入擇時策略試圖進一步改善其績效。為此，研究參考卓育辰(2021)所提出之量化選股模型，並針對架構重新設計，使其具備在不同國家轉換回測之機制，除了實驗財務因子在現今美股的選股效果外，亦加入多項技術策略以驗證是否能夠一致性的改善報酬或風險。 本研究透過 Python 建構回測系統，使系統在擴充上更具彈性。系統使用七項單因子及三項雙因子以實驗財務因子在現今美國市場上的選股效果，並使用買入持有、逆勢布林通道、逢低買進及逆勢高低通道四項技術策略驗證是否能夠對績效帶來一致性的影響。除了財務因子與技術策略兩項參數外，系統亦提供兩項市場、三種最佳化窗格、兩種資金配置法、十個群組與五種持有股數等五種可調參數，以實驗不同配置對績效的影響。 實驗結果發現有些財務因子在現今的美國市場仍存在篩選股票的價值，其中市值營收比在相關性以及獲利性上的表現較為良好，而技術策略的使用普遍能夠降低最大交易回落，為投資人帶來較低的投資風險。 ;Many scholars have attempted to find the factors that have a pattern of performance and bring profit to investors from an empirical perspective by quantifying multi-factor models. In addition to factor stock selection, some literature tried to combine timing strategy to improve investment performance. Therefore, this research refers to the quantitative selection model proposed by Zhuo, Yu-Chen and redesign the architecture. Make the model have a mechanism for switching between countries. To verify stock selection performance of the financial factor on U.S. stocks and also testing weather technical strategy could improve profit or risk consistency. This research’s backtesting system is built in Python to make the system more flexible in terms of expansion. The system uses seven single-factor and three dual-factor to test the effectiveness of financial factor stock selection in today′s U.S. stock. It also using buy-and-hold, inverse Bollinger Bands, bargaining hunting, and inverse high-low channels to test whether the technical strategies have a consistent impact on performance. In addition to the factor and technical strategy, the system also provides five adjustable parameters, including two markets, three optimization windows, two capital allocation methods, ten groups and five shareholdings, to test the impact of the performance when experiment adjust it allocation. Experiment show that some financial factors are still have value of stock selection in today′s U.S. stock, especially PS have best correlation and profitability. And also found that technical strategies could generally reduces maximum drawdown, bring the lower investment risk for investors.
