本研究從台灣50成分股中，篩選出44家公司當作樣本。蒐集2007-2019年間的股價資料，以技術指標當作模型的輸入變數，應用卷積神經網路、長時間短期記憶模型於投資策略上，並結合凱利法則配置投資組合權重。實證結果發現長時間短期記憶模型在訓練期間(2007-2015)、測試期間(2016-2019)內預測股票漲跌準確率表現皆比卷積神經網路優異。實證結果也顯示使用長時間短期記憶模型建構之策略相比元大台灣50 ETF績效，各年度夏普值大多數表現得比元大台灣50 ETF優異。顯示使用深度學習與凱利法則在投資策略上，可以在控制風險的前提下，得到不錯的策略績效。

This Research selects 44 companies from constituent stocks of Taiwan 50 Index as a sample. Collect stock price data from 2007 to 2019 and use technical indicators as input variables of the model, then apply Convolutional Neural Networks、Long Short Term Memory Network to investment strategies. In this research, Kelly criterion is used to allocate stock weights. Empirical results show that Long Short Term Memory Network performs better than Convolutional Neural Network in the accuracy of predicting stock movement during the training period (2007-2015) and the test period (2016-2019). Empirical results also show that most of the annual Sharpe ratios of portfolios constructed by Long Short Term Memory Network are greater than that of Yuanta Taiwan 50 ETF. In the end, this research shows that using deep learning method and Kelly criterion in portfolio construction can get good performance on the premise of controlling risks.

[1] Adebiyi, A. A., Adewumi, A. O., & Ayo, C. K. (2014). Comparison of ARIMA and artificial neural networks models for stock price prediction. Journal of Applied Mathematics, 2014, 1-7.
[2] Asness, C. S., Ilmanen, A., Israel, R., & Moskowitz, T. J. (1998). Investing with style. Journal of Investment Management, 13(11), 27-63.
[3] Byrnes, T., & Barnett, T. (2018). Generalized framework for applying the Kelly criterion to stock markets. International Journal of Theoretical and Applied Finance, 21(5), 1-13.
[4] Bai, S., Kolter, J. Z., & Koltun, V. (2018). An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. Retrieved April 11, 2020, from https://arxiv.org/abs/1803.01271.
[5] Black, F., & Litterman, R. (1991). Asset allocation: combining investor views with market equilibrium. The Journal of Fixed Income, 1(2), 7-18.
[6] Black, F., & Litterman, R. (1992). Global portfolio optimization. Financial Analysts Journal, 48(5), 28-43.
[7] Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14, 179-211.
[8] Hoseinzade, E., & Haratizadeh, S. (2018). CNNpred: CNN-based stock market prediction using a diverse set of variables. Expert Systems with Applications, 129, 273-285.
[9] He, G., & Litterman, R. (1999). The intuition behind Black-Litterman model portfolios. New York, NY: Goldman Sachs.
[10] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
[11] Kelly, J. L. (1956) A new interpretation of information rate. Bell System Technical Journal, 35(4), 917-926.
[12] Kwon, Y. K., & Moon, B. R. (2007). A hybrid neurogenetic approach for stock forecasting. IEEE Transactions on Neural Networks, 18(3), 851-864.
[13] LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., & Jackel, L. D. (1989). Backpropagation applied to handwritten zip code recognition. Neural Computation, 1, 541-551.
[14] Markowitz, H. (1952). Portfolio selection. The journal of finance, 7(1), 77-91.
[15] Nelson, D. M. Q., Pereira, A. C. M., & De Oliveira, R. A. (2017). Stock market's price movement prediction with LSTM neural networks. Proceedings of 2017 International Joint Conference on Neural Networks, 1419-1426. doi: 10.1109/IJCNN.2017.7966019
[16] Ohlsson, E., & Markusson, O. (2017). Application of the Kelly criterion on a self-financing trading portfolio - An empirical study on the Swedish stock market from 2005-2015. Retrieved April 11, 2020, from https://reurl.cc/oLb6Xg.
[17] Ross, S. (1976). The arbitrage theory and capital asset pricing. Journal of Economic Theory, 13, 341-360.
[18] Rounaghi, M. M., & Zadeh, F. M. (2016). Investigation of market efficiency and financial stability between S&P 500 and London stock exchange: Monthly and yearly forecasting of time series stock returns using ARMA model. Physica A: Statistical Mechanics and its Applications, 456, 10-21.
[19] Sharpe, W. F. (1964). Capital asset prices: a theory of market equilibrium under conditions of risk. Journal of Finance, 19(3), 425-442.
[20] Sharpe, W. F. (1992). Asset allocation: management style and performance management. Journal of Investment Management, 18(2), 7-19.
[21] Wu, M. E., & Chung, W. H. (2018). A novel approach of option portfolio construction using the Kelly criterion. IEEE Access, 6(1), 53044-53052.
[22] Zhai, Y., Hsu, A., & Halgamuge, S. K. (2007). Combining news and technical indicators in daily stock price trends prediction. In D. Liu, S. Fei, Z. Hou, H. Zhang & C. Sun (Eds.), Advances in Neural Networks – ISNN 2007 (pp. 1087-1096). Berlin, Germany: Springer.