Prediction model for stock market using news based different Classification, Regression and Statistical Techniques: (PMSMN)

Abstract

The proposed work in the term is focused on developing the prediction model on crude commodity based on its price movement due to news released by various sources. Further the model is improved by applying different computing techniques. The primary objective is to derive model for investment decision for crude commodity. The decision strategy would be driven by analysing stock price fluctuation based on sector preference and news released. Comparison and analysis of forecasting techniques for model prediction requirements. Deriving the precise forecasting technique by combining existing prediction approaches. Analysing relevant released news by text classification and impact calculation strategies. Deciding the prediction method to be use in model by performance comparison of following prediction techniques: Regression Modelling techniques, Classification Techniques, Statistical Techniques. Compare performance parameters like Throughput, accuracy and Error Rate in proposed model.