딥러닝을 활용한 자산분배 시스템

Portfolio System Using Deep Learning

  • 김성수 (숭실대학교 글로벌미디어학부) ;
  • 김종인 (숭실대학교 글로벌미디어학부, (주)Fait) ;
  • 정기철 (숭실대학교 글로벌미디어학부)
  • 투고 : 2018.12.14
  • 심사 : 2019.02.12
  • 발행 : 2019.02.28


딥러닝 네트워크 기반의 알고리즘의 발전으로 인공지능은 전세계적으로 빠른 성장세를 보이고 있다. 그 중 금융은 인공지능이 가장 많이 활용될 분야로 예상되고 있으며 최근 많은 연구가 되고 있다. 기존의 딥러닝을 사용한 재무 전략은 단일 종목에 대한 주가 예측에만 치중되어 있어 변동성에 취약하다. 따라서 본 연구는 딥러닝을 이용하여 펀드 구성 종목을 산출하고 종목들을 분산투자하여 ETF 상품을 구성하는 모델을 제안한다. 실험 결과로 제안하는 모델을 통해 코스피 100 지수를 대상으로 하는 성능을 분석하며 수익률 또는 안정성 측면에서 향상된 결과를 확인하였다.

As deep learning with the network-based algorithms evolve, artificial intelligence is rapidly growing around the world. Among them, finance is expected to be the field where artificial intelligence is most used, and many studies have been done recently. The existing financial strategy using deep-run is vulnerable to volatility because it focuses on stock price forecasts for a single stock. Therefore, this study proposes to construct ETF products constructed through portfolio methods by calculating the stocks constituting funds by using deep learning. We analyze the performance of the proposed model in the KOSPI 100 index. Experimental results showed that the proposed model showed improved results in terms of returns or volatility.


SOJBB3_2019_v24n1_23_f0001.png 이미지

Fig. 1 Autoencoder Model Structure

SOJBB3_2019_v24n1_23_f0002.png 이미지

Fig. 2 Deep Learning Portfolio Process

SOJBB3_2019_v24n1_23_f0003.png 이미지

Fig. 3 Portfolio Selection Model

SOJBB3_2019_v24n1_23_f0004.png 이미지

Fig. 4 Graphs of Funds Return

Table 1.1 Result of Experiment(30 stocks)

SOJBB3_2019_v24n1_23_t0001.png 이미지

Table 1.2 Result of Experiment(50 stocks)

SOJBB3_2019_v24n1_23_t0002.png 이미지

Table 1.3 Result of Experiment(100 stocks)

SOJBB3_2019_v24n1_23_t0003.png 이미지

Table 2 ETF Components(30)

SOJBB3_2019_v24n1_23_t0004.png 이미지


  1. Shim, H. and Kim, K., "A Business Model Using Artificial Intelligence (AI) of Korean Companies," KITA, TRADE FOCUS, Vol. 2, No. 97, 2018.
  2. "Status of Machine Learning at Home and Abroad in the Field of Finance," FSI, 2017.
  3. KEB Hana Bank, "2018 Korean Roboadvidser Report," KEB Hana Bank HAI Robo Center, 2018.
  4. Hwang, R., Kim, S., Lee, D., and Nam D., "A Directional Distance Function Approach on the Efficiency of Chinese Commercial Banks," Journal of the Korea Industrial Information Systems Research, Vol. 17, No. 2, pp. 81-94, 2017.
  5. Ban, J., Kim, M., and Jeon, Y., "Search Frequency in Internet Portal Site and the Expected Stock Returns," Journal of the Korea Industrial Information Systems Research, Vol. 21, No. 5, pp. 73-83, 2016.
  6. Kim, S. and Ahn, H., "Development of an Intelligent Trading System Using Support Vector Machines and Genetic Algorithms," Journal of Intelligence and Information System, Vol. 16, No. 1, pp. 71-92, 2010.
  7. Lee, W., "A Deep Learning Analysis of the KOSPI's Directions," Journal of the Korean Data and Information Science Society, Vol. 28, No. 2, pp. 287-295, 2017.
  8. Kim, Y., Shin, E., and Hong, T., "Comparison of Stock Price Index Prediction Performance Using Neural Networks and Support Vector Machine," The Journal of Internet Electronic Commerce Resarch, Vol. 4, No. 3, p p. 221-243, 2004.
  9. Kim H., Kim K., and Jeong D., "A Study on the Price Determination of KOSPI 200 Futures using Artificial Neural Network Model," Korea Insurance Research Institute, Insurance Financial Research, Vol. 13, No. 3, pp. 155-176, 2003.
  10. Joo, I., Choi, S., "Stock Prediction Model based on Bidirectional LSTM Recurrent Neural Network," Journal of Korea Institute of Information, Electronics, and Communication Technology, Vol. 11, No. 2, pp. 204-208, 2018.
  11. Shin, D., Choi, K., and Kim, C., "Deep Learning Model for Prediction Rate Improvement of Stock Price Using RNN and LSTM," Journal of Advanced Information Technology and Convergence, Vol. 15, No. 10, pp. 9-16, 2017.
  12. Kim, S., Hong, K., "Development and Performance Analysis of Predictive Model for KOSPI 200 Index using Recurrent Neural Networks," Journal of the Korea Industrial Information Systems Research, Vol. 22, No. 6, pp. 23-29, 2017.
  13. Heaton, J. B., Polson, N. G., and Witte, J. H., "Deep Portfolio Theory," University of Chicago, 2016.
  14. Ryu, J., Shin, H., "Portfolio Selection Strategy Using Deep Learning," Journal of Information Technology and Architecture, Vol. 15, No. 1, pp. 43-50, 2018.