A local and global event sentiment based efficient stock exchange forecasting using deep learning

Haider Maqsood, Irfan Mehmood, Muazzam Maqsood, Muhammad Yasir, Sitara Afzal, Farhan Aadil, Mahmoud Mohamed Selim, Khan Muhammad

Abstract: Highlights•A deep learning-based method is proposed to forecast stock prices for top companies from four countries that were selected from developed, underdeveloped and emerging economies.•We investigate the effect of different famous events occurred from 2012 to 2016 comprising 11.42 Million tweets on stock prediction.•The effect of local and global events for stock companies from each country has been investigated.•The results show that event sentiment improves the results for stock forecasting depending upon local and global events.