IBOVESPA index volatility forecasting using neural networks
Author: Gabriel Ogawa Cruz
Supervisor: Roberto Hirata Jr.
Abstract: Volatility forecasting is a core part of risk management and options pricing in the stock exchange context. While using time series models is still common for such forecasting, this work will explore the accuracy of using a long short term memory (LSTM) recurrent neural network (RNN) for forecasting the volatility of Brazil's stock price index, IBOVESPA.