Fast Gibbs sampling for the local and global trend Bayesian exponential smoothing model
This work addresses a practical bottleneck for users of time series forecasting models by making a high-performing method more computationally efficient, though it is incremental as it builds directly on prior work.
The paper tackled the computational expense of a state-of-the-art Bayesian exponential smoothing model by proposing modifications and a bespoke Gibbs sampler, improving sampling time by an order of magnitude while maintaining competitive or superior accuracy on the M3 dataset.
In Smyl et al. [Local and global trend Bayesian exponential smoothing models. International Journal of Forecasting, 2024.], a generalised exponential smoothing model was proposed that is able to capture strong trends and volatility in time series. This method achieved state-of-the-art performance in many forecasting tasks, but its fitting procedure, which is based on the NUTS sampler, is very computationally expensive. In this work, we propose several modifications to the original model, as well as a bespoke Gibbs sampler for posterior exploration; these changes improve sampling time by an order of magnitude, thus rendering the model much more practically relevant. The new model, and sampler, are evaluated on the M3 dataset and are shown to be competitive, or superior, in terms of accuracy to the original method, while being substantially faster to run.