透過您的圖書館登入
IP:18.116.239.195
  • 學位論文

使用卷積神經網路預測台灣加權指數

Forecasting Taiwan Capitalization Weighted Stock Index by Using Convolutional Neural Network

指導教授 : 袁賢銘

摘要


近年來,深度學習被廣泛的應用於各個領域,而金融預測是其中一個熱門的應 用。卷積神經網路是深度學習中的一種經典的模型,且擅長於提取出圖片中的 各種特徵。在本篇論文中,我們使用了來自 Google 的 Chollet, F 所提出的卷積 神經網路,Xception 來預測台灣加權指數。此外,我們使用未來 20 天、30 天 和 40 天的中位數為依據來標記我們的圖片。並且,基於這些預測的結果,我們 提出了一種方法可以找出相對於他們的最佳交易策略。而在我們使用找到的交 易策略去模擬交易台灣指數的 ETF,如 0050、0056、00692 和 006208 後,所得 到的結果顯示,我們所提出的交易策略可以獲得較多的收益,相較於傳統的交 易策略,像是買進並持有,或是一些技術指標。

並列摘要


Deep learning has been widely used in many research areas recently. One of the common applications is financial forecasting. Convolutional neural network, a class of deep learning, which is capable of capturing complex features from the images. In this paper, we proposed a method for forecasting Taiwan capitalization weighted stock index by using Xception, a convolution neural network presented by Chollet, F from Google. Furthermore, with the labeling technique we proposed which aims to predict the median of the closing price of future 20, 30, or 40 days. Based on the results produced by each model, we propose a method to find the optimal trading strategy. Then, after the simulation based on trading the common ETFs in Taiwan which includes 0050.TW, 0056.TW, 00692.TW, and 006208.TW, the promising results show that we can make more profits than those traditional trading strategies such as Buy and Hold and some technical indicators.

參考文獻


1. Regnault, J., Calcul des chances et philosophie de la bourse. 1863: Mallet-Bachelier.
4. Wilder, J.W., New concepts in technical trading systems. 1978: Trend Research.
6. Appel, G., Technical analysis: power tools for active investors. 2005: FT Press.
12. Hyndman, R.J. and G. Athanasopoulos, Forecasting: principles and practice. 2018: OTexts.

延伸閱讀