應用基因演算法達成知識之最適化--以台灣股市技術分析指標為例

Abstract

隨著資訊科技的日新月異, 用以研判各種金融商品買賣時機之交易系統, 也日益繁多。 這多起因於在瞬息萬變的市場中, 自動化的交易系統可以 協助吾人掌握先機之故。 在過去, 多數交易系統的知識庫是依據傳統的 統計經濟模型所建立。而今日人工智慧領域的長足發展, 為吾人提供了更 多的選擇。其中,基因演算法藉著極有效率的空間搜尋能力, 能迅速將影 響因子歸納出, 因此頗適合用來作為交易策略知識的搜尋機制。因此，本 研究嘗試提出一個以基因演算法為核心架構之知識庫模組的建構系統。首 先，將知識庫中的規則模組語法，加以定義；再藉由基因演算法的搜尋機 制，從資料庫中擷取影響投資決策的技術分析指標，透過系統模擬的過程 加以運算，求得最佳的交易策略知識。於是一個完整的交易系統模型便可 以產生。最後，由本研究的結果,可以結論出：以基因演算法來建構交易 系統的知識庫，可以有效的將關鍵的規則集納入。包括關鍵因子與規則集 所需的參數，皆可一併求得。而經過實際的模擬驗證後，由此產生的規則 集，在獲利能力與風險管理的控制上，也都有優異的表現。 The progress of computer information systems has made tremendous growth of trading systems. The major benefit gained from trading systems is that it can response to markets quickly. In the past decade, most knowledge-based of trading systems are constructed by statistical or economic models.Recently, the progress of artificial intelligence technique has created more methods to achieve this goal. Genetic algorithms are one of these AI-based methodology. The excellent ability of searching solutions has made them quite suitable for constructing the knowledge bases. The major objective of this research is to propose a model to construct knowledge bases of trading systems by genetic algorithms. First of all , both of the knowledge representation method and grammars needed for trading rules will be defined. The second step is that the trading rules will be generated by genetic algorithms. And in the third stage , these rules will be evaluated by the pre-defined fitness function and the qualified rules will be extracted. Finally, a complete trading systems along with the qualified rules would be demonstrated.In the end, according to the results of this research, the conclusion can be made as follows: using genetic algorithms to construct the knowledge base of the trading system is particularly efficient in finding out the key rule models, including the necessary parameters of the rule models. And the simulation result shows that the performance of these rules are quite excellent in both profit margin and risk control.