A new BRB model for technical analysis of the stock market

Abstract

To predict the trend of stock prices, a belief rule base (BRB) assessment model based on different technical indicators is proposed in this paper. The proposed BRB-based model includes three BRBs, denoted as BRB_1, BRB_2 and BRB_3. BRB_1 is used to capture the relationship between the price trend of the moving average (MA) and buy/sell decisions. BRB_2 is used to investigate the conditions of moving average convergence and divergence (MACD). BRB_3 is employed to represent the stochastic indicator (KD) states. The above three indicators are commonly used in stock analysis, and they usually need to be used together to achieve a more accurate analysis of the stock price trend. In the BRB model, the initial values of some parameters are provided by experts to construct the elementary algorithm logic, but these are unlikely to result in an accurate assessment. Therefore, on the basis of the maximum likelihood (ML) algorithm, an optimal algorithm for training the parameters of the assessment model is further proposed. Taking the trend of the Chinese stock market as the research object, an average MSE of 0.3242 is obtained using this model. The results indicate the potential application of the proposed model in the financial industry.

Keywords

• Moving average
• Moving average convergence and divergence
• Stochastic indicator
• Belief rule base (BRB)
• Chinese stock market