Multi model-Based Hybrid Prediction Algorithm (MM-HPA) for Stock Market Prices Prediction Framework (SMPPF)

In financial arena, stock markets have influence on the performance of organizations and investors. Stock markets are highly dynamic in nature, and predicting the stock prices is a challenging task. Reliable prediction needs a superior method for drawing inferences from the data available. In machine learning domain, artificial neural networks are considered as accurate prediction models but using these techniques to forecast the stock market price up to some extent. So, there is a need to improve the accuracy of the model. Towards this end a framework known as Stock Market Prices Prediction Framework (SMPPF) is proposed. This framework has the underlying algorithm known as Multi-Model based Hybrid Prediction Algorithm (MM-HPA) that is the combination of linear and non-linear models including Genetic Algorithm (GA). It is the combination of linear and nonlinear models including genetic algorithm (GA). In linear model, use autoregressive moving average model and in nonlinear model, use the deep learning model known as recurring neural network. Finally apply the GA for finding optimal parameters for the hybrid model. The proposed framework is evaluated with a prototype built on Python data science platform. The empirical results revealed that the difficulty of traditional models in capturing patterns from nonlinear data is overcome by the proposed hybrid model.