技術分析在期貨與股票交易市場中，大量地被運用。其中交易程式在近年大幅地被運用，因人為分析的市場交易，常會被心理因素影響，而錯過最佳的買賣時機，因此，程式交易在近年來被大幅地運用。過去程式交易績效的穩定性分析，都是透過觀察交易平台所產生的參數最佳化圖表，來進行人工分析、比對，但其過程繁雜，且在高維度時難以比較。因此，本研究透過多維度資料分析的方法，發展出一套程式交易的參數最佳化績效高原自動尋找平台，希望藉此協助使用者透過本研究，自動化地找到參數最佳化績效高原。同時，本研究也利用維度堆疊的方法，提供一套便捷的多維度參數最佳化績效高原視覺化驗證工具。 本研究也將使用實際的程式交易策略最佳化資料，以及可推展至多維度的數學模型來生成樣本資料，驗證本系統的正確性。 本研究期望透過傳統分析程式交易穩定性的方法加以改良，發展出一套自動化的交易程式最佳化的多維度分析平台，供投資大眾使用。;Technical analysis is widely used in the futures and stock markets. Program trading is used substantially in recent years, because artificial analysis of market transactions, often influenced by psychological factors, and miss the best time to sell, therefore, program trading has been heavily used in the recent years. Stability analysis of program trading performance in the past, it is done by observing the parameter optimization charts from the trading platform, in order to analysis, and alignment. But the process is complicated, and difficult to compare when facing higher dimension. Therefore, this study developed a parameter optimizing performance plateau automatically finding platform for program trading parameter optimization through multi-dimensional data analysis methods, hoping to help users find parameter optimizing performance plateau automatically through this study. Meanwhile, this study also uses the dimensional stacking method, to provide a convenient set of multi-dimensional parametric performance plateau visual verification tool. This study will use the actual program trading strategies optimizing data, and a mathematical model which can be extended up to multi-dimensions to generate sample data, to verify the correctness of the proposed system. This study aims to develop an automated multi-dimensional analysis platform of program trading optimization, through the improvement of the traditional program trading stability analysis methods, for the investing public.
