Hong Kong as a financial center have high frequency of trading every working days. The fluctuation of stock prices is one of concerning issues for investors. It seems that there is no any trackable information and factors affecting the variations of stock price. However, It has been shown that the financial market is “informationally efficient” [Fama, 1965] — stock prices reflect all known information, and the price movement is in response to news or events. In the world, there are many companies trying to use deep learning to predict the movement of stocks. Some has a great success and stride to an advanced step to use deep learning to predict the movement of stock. Certainly, there are also some that are frustrated on the prediction of stock markets due to random and stochastic reason of stock market. However, Hong Kong, an international hub, still lags many countries behind using deep learning to predict the movement of stock prices and there is a small amount of researches related to this topic. Hence, this paper will focus on how to make use of deep learning this modern techniques to predict the stock prices in a stochastic environment and how to study the characteristic of Hong Kong stock market with the aid of deep learning.