Stock market prediction is the most dynamic problems in this era. It is affected by political decisions, policy decisions of the companies, rumors, conversations in the public forums, demand and supply, investor’s mood, weather etc. Stock market depends on various attributes like opening price, closing price etc. Many machine learning approaches, time series analysis methods, technical analysis methods were existing, but still there is lots of scope for a new mechanism to predict the stock market variations. We propose a novel classifier named Hybrid Fuzzy Cellular Automata to predict the stock market variations. This classifier is trained and tested with BSE data to perform stock predictions. This classifier output is compared with the existing standard algorithms. The prediction rate was considerably increased by 1.6%.

KEYWORDS:  Cellular Automata, Stock Prediction, Dynamic Data.