Stock price prediction Based on Generative Adversarial Network

As the "barometer" of economic development, the stock market occupies an increasingly important position in the financial field. The prediction of stock price has always been an important subject for scholars. Faced with the impact of Internet information, more and more investors make comments on various social media platforms, which will imperceptibly affect investors investment decisions, and will also have an impact on stock market fluctuations. Therefore, this paper proposes a multi-factor stock price prediction model based on generative adversarial networks from the perspective of stock review text mining. Experimental results show that our GAN has good performance in stock close price prediction when compared to other statistical models and machine learning models.