Stefan-Razvan ANTON, Octavian POSTAVARU and Antonela TOMA
Center for Research and Training in Innovative Techniques of Applied Mathematics in Engineering, University Politehnica of Bucharest, Bucharest, Romania
As our world moves more and more into the online medium so does our data. This massive movement allows the analysis of financial markets using methods that would not have been possible 10 years ago due to a lack of sufficient data. In this paper, we review some basic ideas of stock market data handled as a time series, need of RNN (Recurrent Neural Network), survey previous works, and use a LSTM (Long Short-Term Memory) network to forecast stock value during and after the 2020 market crash. The prediction accuracy is calculated and investigated with reference to S&P 500, PHLX Semiconductor (SOX), and XLRE.