Stock price prediction using Generative Adversarial Networks

Deep learning is an exciting topic. It has been utilized in many areas owing to its strong potential. For example, it widely used the financial area which vital society, such as high-frequency trading, portfolio optimization, fraud detection and risk management. Stock market prediction one of most popular valuable finance. In this paper, proposes a stock model using Generative Adversarial Network (GAN) with Gated Recurrent Units (GRU) generator that inputs historical price generates future Convolutional Neural (CNN) discriminator discriminate between real generated price. Different from traditional methods, limited forecasting on one-step-ahead only, by contrast, deep algorithm possible conduct multi-step ahead more accurately. study, chose Apple Inc. closing target price, features S&P 500 index, NASDAQ Composite U.S. Dollar etc. addition, FinBert generate news sentiment index for additional predicting feature. Finally, paper compares proposed GAN results baseline model.