本研究使用深度學習中的卷積神經網路，針對美國資產規模前十大ETF來做漲跌趨勢的預測，建立卷積神經網路架構，再利用收集的2000年至2018年的歷史資料來做資料的前處理、訓練模型，把得到的模型進一步去預測未來的漲跌情況。

本研究在訓練資料中除了ETF歷史資料以外，還選擇了多面向的技術指標、包含常見的移動平均線、相對強弱指數以及其他有關之技術分析。而集體思維指的是群體決策中的一種現象，在本研究中指的是群眾預期未來市場變化進而做出的反應，是以VIX指數(俗稱恐慌指數)作為表現。本研究對於卷積神經網路模型應用在預測ETF漲跌趨勢提供了兩種不同的資料標籤方式所建構的模型，其投資實驗也證明了利用此種方法可以獲得不錯的年化收益率(25.58%)及較高的上漲猜對的機率(82.04%)，此種方法建立之模型相比於傳統的買入持有策略、隨機買入策略表現的結果都更好，顯示本研究的實驗結果在投資上擁有更好的效果，輔助投資人在投資時作為參考。

With the growth of computer hardware speed, artificial intelligence, which requires a lot of computing technology, is popular again. Due to the progress of GPU performance, effective parallel computing accelerates the operations required by the algorithm, allows these artificial intelligence technology being more convenient and effective in different fields. Deep learning is one of the artificial intelligence that has been discussed by many people in recent years.

This study is based on the convolutional neural network in deep learning, and forecasts the ups and downs of the US Top 10 ETF. First, the convolutional neural network architecture of the each ETF is established. Then, the historical data from 2000 to 2018 will be preprocessed to train the model, and further, the model will be used to predict the future trend.

In addition to ETF historical data, this study selected multi-oriented technical analyses, including simple moving average, relative strength index and other related technical analyses. Groupthink is a psychological phenomenon that occurs within a group of people in which the desire for harmony or conformity in the group results in an irrational or dysfunctional decision-making outcome. In this study, groupthink means the reaction of the stockholder in anticipation of future market changes that is represented the VIX index (commonly known as Volatility Index). This study provides multiple sets of parameters for predicting ETF ups and downs in convolutional neural network models. The experimental performance also proves that this method can obtain a good return rate (25.58%) and provide investors as a reference in the future.

[1] 尤韻涵，(2009)。台股指數開盤價格之預測-應用類神經網路及灰預測模型，輔仁大學，經濟學研究所，台北。
[2] 吳哲緯，(2017)。使用深度學習卷積神經網路預測股票買賣策略之分類研究，國立中山大學，資訊管理學系研究所，高雄。
[3] 林婉茹，(2004)。類神經網路於台灣50指數ETF價格預測與交易策略之應用，輔仁大學，金融研究所，台北。
[4] CS231n Convolutional Neural Networks for Visual Recognition, Retrieved June 12 2019, from: http://cs231n.github.io/convolutional-networks/
[5] Dingli, A., & Fournier, K. S. , (2017). Financial time series forecasting-a machine learning approach, Machine Learning and Applications: An International Journal, 4(1/2), 3., 11-27.
[6] Ding, X., Zhang, Y., Liu, T., & Duan, J., (2015). Deep Learning for Event-Driven Stock Prediction, Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI 2015), 2327-2333.
[7] Gudelek, M. U., Boluk, S. A., & Ozbayoglu, A. M., (2017). A deep learning based stock trading model with 2-D CNN trend detection, In 2017 IEEE Symposium Series on Computational Intelligence (SSCI), 1-8.
[8] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition, In Proceedings of the IEEE conference on computer vision and pattern recognition(CVPR), 770-778.
[9] Kimoto, T., Asakawa, K., Yoda, M., & Takeoka, M., (1990). Stock market prediction system with modular neural networks, In 1990 IJCNN international joint conference on neural networks, 1-6.
[10] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks, In Advances in neural information processing systems, 1097-1105.
[11] Kuo, R. J., Chen, C. H., & Hwang, Y. C., (2001). An intelligent stock trading decision support system through integration of genetic algorithm based fuzzy neural network and artificial neural network, Fuzzy sets and systems, 118(1), 21-45
[12] LeCun, Y. and Y. Bengio, (1995). Convolutional networks for images, speech, and time series. The handbook of brain theory and neural networks. The handbook of brain theory and neural networks, London, England, 255-258.
[13] LeCun, Y., L. Bottou, Y. Bengio, and P. Haffner, (1998). Gradient-based learning applied to document recognition, Proc. IEEE, vol. 86, no. 11, 2278-2324.
[14] Lu, T. H. and Y. C. Chen, (2015). Trend definition or holding strategy: What determines the profitability of candlestick charting? Journal of Banking & Finance, Volume 61, 172-183.
[15] Luca Di Persio, Oleksandr Honchar. (2016) Artificial neural networks approach to the forecast of stock market price movements, International Journal of Economics and Management Systems, 1, 158-162.
[16] Moghaddam, A. H., Moghaddam, M. H., & Esfandyari, M. , (2016). Stock market index prediction using artificial neural network, Journal of Economics, Finance and Administrative Science, 21(41), 89-93.
[17] One by One [1 x 1] Convolution - counter-intuitively useful, Retrieved June 12 2019, from: https://iamaaditya.github.io/2016/03/one-by-one-convolution/
[18] Review: GoogLeNet (Inception v1)— Winner of ILSVRC 2014 (Image Classification), Retrieved June 12 2019, from: https://medium.com/coinmonks/paper-review-of-googlenet-inception-v1-winner-of-ilsvlc-2014-image-classification-c2b3565a64e7
[19] Sezer, O. B., Ozbayoglu, A. M., & Dogdu, E., (2017). An artificial neural network-based stock trading system using technical analysis and big data framework, Proceedings of the SouthEast Conference on - ACM SE 17, 223-226.
[20] Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A., (2015). Going deeper with convolutions, In Proceedings of the IEEE conference on computer vision and pattern recognition, 1-9.
[21] Vargas, M. R., Beatriz S. L. P. De Lima, & Evsukoff, A. G. , (2017). Deep learning for stock market prediction from financial news articles, 2017 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA), 60-65.
[22] What-is-the-VGG-neural-network, Retrieved June 12 2019, from: https:// www.quora.com/What-is-the-VGG-neural-network