A1765
Title: Combining econometrics and machine learning to forecast realized volatility of exchange rates
Authors: Aleksandr Pereverzin - University of East Anglia (United Kingdom) [presenting]
Abstract: Time series of financial volatility is well known for having a complex structure including several heterogeneous patterns: linear and nonlinear, long-run and short-run, etc. We propose a new two-component model of realized volatility that is based on methodological combination of econometric and machine learning approaches. In our model, Autoregressive Fractionally Integrated Moving Average (ARFIMA) framework is used to capture the linear component of realized volatility while the artificial neural network is used to model the corresponding nonlinear part. We also develop a modification of the cyclical volatility model where artificial neural networks are used to model both trend and cyclical components of realized volatility. The proposed models provide an improvement in out-of-sample forecasting accuracy over the competing approaches.
