在股票市場中,每天皆有許多投資者嘗試以許多不同的方法預測股票價格。傳統上,基本分析與技術分析是兩種較常見的股價預測工具,近年來時間數列分析法已逐漸被用來作為股價預測。為了能更準確預測未來股價趨勢,本研究提出一套改良式類神經網路預測模式用以預測股價,並與三種時間數列方法(ARIMA、灰預測與類神經網路)進行比較。本研究所提出的改良式類神經網路預測模式將應用於台灣股票市場上市公司航運類股中三家主要公司股價預測。實證結果顯示四種預測模式中,本研究所提出之改良式類神經網路預測的預測績效最佳。本研究所提出的改良式類神經網路預測模式將可有效提供相關人員作為未來股價預測使用。
Many investors try their best to forecast stock prices accurately. Accurate forecasting stock prices has been regarded as one of the most important issues in time series analysis. In this research, an improved ANNs model is proposed to forecast the stock prices in Taiwan. Daily stock price data from three companies is used in this research. The mean absolute percentage error (MAPE) is used to compare the performance of the improved neural networks model against three other models (i.e., the ARIMA model, the ANNs model and the GM(1,1) model). Results show that the improved neural networks model outperforms among the four models.