應用具泰勒級數概念之類神經網路建構台灣股價趨勢預測模式

Abstract

股票投資分析主要有基本分析與技術分析二種方法。基本分析是假設股票皆有合理價格的存在，而公司的股價會隨著合理價格來做調整。技術分析方法則是著重在股票過去的成交價與成交量，利用圖表、數據分析預測整個股市或個股未來的價格變化。

電子股一直是投資人的最愛，在大盤扮演主流股的地位。然而，近年來電子股逐漸失去大盤主流股的地位。塑化、鋼鐵、航運、金融等類股在大盤的地位日益重要。Black與Scholes於1973年提出選擇權定價理論，選擇權的價格受到股價、時間與相關變數所影響，藉由調整適當的避險比例，構成一無風險之投資組合；Black-Scholes偏微分方程式結構與泰勒級數相當類似。本論文嘗試以市場上經常使用的基本分析變數每股盈餘及技術分析指標MACD建立符合泰勒級數精神之類神經網路模型，對台灣股市各類股股價走勢加以探討。選取具產業代表性的公司如台塑、燁輝、長榮、彰銀等做訓練及測試，藉由實證研究，找出合理的股價走勢分析技巧，並以此來研判股價未來走勢及決定買賣股票的時機。實證結果顯示具泰勒級數精神之類神經網路股價趨勢預測模式與以原始量建構之股價趨勢預測模式比較，顯著有較高的準確率及獲利率。

Stock market is nonlinear and semi-structured. In other words, there are many different situations and the environment states are changing quickly. In Taiwan, stock market is always affected by political factors. So the fluctuation of stock price is always larger than other country. Therefore, to predict the trend of stock price is more important. Traditional trading strategy like regression model and random walk are limited in fixed time interval and can not perform well. There are many researches done on predicting stock price, most of them are using Neural Network as its prediction technique, but their model only use raw data. This paper adopts Taylor series concept based Neural Network to predict the trend of Taiwan stock price, which uses fundamental analysis indicator and technical analysis indicator based on Taylor series formed data. This Neural Network model can perform well in stock price prediction and better than raw data based model. The simulation results showed that this system could get an outstanding trading profit and accuracy rate.