Neural Generalised AutoRegressive Conditional Heteroskedasticity
This work addresses the challenge of capturing dynamic volatility in financial markets for practitioners and researchers, representing an incremental improvement by adapting existing models with neural networks.
The authors tackled the problem of modeling conditional heteroskedasticity in financial time series by proposing Neural GARCH, a neural network adaptation of traditional GARCH models with time-varying coefficients parameterized by a recurrent neural network, and found that the Neural Students t variant consistently outperformed others on a wide range of univariate and multivariate financial datasets.
We propose Neural GARCH, a class of methods to model conditional heteroskedasticity in financial time series. Neural GARCH is a neural network adaptation of the GARCH 1,1 model in the univariate case, and the diagonal BEKK 1,1 model in the multivariate case. We allow the coefficients of a GARCH model to be time varying in order to reflect the constantly changing dynamics of financial markets. The time varying coefficients are parameterised by a recurrent neural network that is trained with stochastic gradient variational Bayes. We propose two variants of our model, one with normal innovations and the other with Students t innovations. We test our models on a wide range of univariate and multivariate financial time series, and we find that the Neural Students t model consistently outperforms the others.