Ascertaining Price Determinants and Forecasting Model for BSE Sensex Stocks Jain Aayush Research Scholar, Institute of Management, Christ University, Bengaluru, India. aayushjain530@gmail.com Online published on 12 May, 2017. Abstract There have been numerous studies carried out that either highlight the determinants of stock price or forecast the future prices, but, few for Indian bourse. The following study is an attempt to depict 360 degrees view to this domain, thus involving 58 predictors categorised into four broad categories: Accounting, Macroeconomic, Technical and International, so that more stress could be laid on variables that actually impact stock price movement while making an investment decision. This study is focused on S&P BSE Sensex companies for the period of 2005–06 to 2015–16, divided into two parts-first focusing on identification of variables that have a significant impact on stock price movement, using Random Forest, whereas second focusing on construction of a model capable of forecasting post-ante annual stock prices, using Multivariate Adaptive Regression Spline (MARS). Accounting variables were found to be most significant, whereas Technical variables were least impactful. The impact of International and Macro-economic variables was found to be unevenly distributed. MARS depicted the presence of white noise in considered dataset and model for stock price prediction at 90.7% accuracy and 0.015 error level. Top Keywords Stock Price, Investment, Forecasting, BSE Sensex. Top |