A0513
Title: A long short-term memory stochastic volatility model
Authors: Nghia Nguyen Trong - University of Sydney (Australia) [presenting]
Minh-Ngoc Tran - University of Sydney (Australia)
Robert Kohn - University of New South Wales (Australia)
David Gunawan - University of New South Wales (Australia)
Abstract: Stochastic Volatility (SV) models are widely used in the financial sector and LongShort-Term Memory (LSTM) models have been successfully used in many large-scale industrial applications of Deep Learning. This present work combines these two techniques in a non-trivial way and proposes a model, called LSTM-SV, for capturing efficiently the dynamics in financial volatility processes. The proposed model overcomes the short-term memory problem in conventional SV models, is able to capture non-linear dependences in the latent volatility process, and often has a better out-of-sample forecast ability. These are illustrated through three financial time series datasets: US stock market index S&P500, Australian stock index ASX200 and Australian-US dollar exchange rates. We argue that there are significant differences in the underlying dynamics between the volatility process of S&P500 and ASX200 datasets and that of the exchange rate dataset. For the stock index data, there is strong evidence of long-term memory and non-linear dependences in the volatility process, while this is not the case for the exchange rates. An user-friendly software is publicly available.
