딥러닝 기법을 활용한 중국 주식시장 예측 연구

Business Analytics is a way of gathering large amounts of data, extracting information, and improving the efficiency of decision making. The global stock market produces large amounts of data every day, and stock prices are influenced by many factors, such as market indicators, economic and industrial conditions. Predicting the direction of stock price fluctuations is an important issue for investors. This study predicts the trend of the stock price in the China Stock Market based on the Deep Learning Method. Five industries were chosen from the China Stock Market, and a total of 82 companies were selected. The five industries are the basic materials, consumer goods, services, healthcare and financial industries. From 2002 to 2015, 3,523 daily stock price data were collected. Twenty-one macroeconomic indicators were included as independent variables. This study predicts stock price using neural network analysis and least squares method. Further studies should be conducted to select key prediction indicators and to improve the accuracy of neural network models.