no code implementations • 25 Feb 2022 • Thabang Mathonsi, Terence L van Zyl
Hybrid methods have been shown to outperform pure statistical and pure deep learning methods at both forecasting tasks, and at quantifying the uncertainty associated with those forecasts (prediction intervals).
1 code implementation • 16 Dec 2021 • Thabang Mathonsi, Terence L. Van Zyl
Difficulties with applying hybrid forecast methods to multivariate data include ($i$) the high computational cost involved in hyperparameter tuning for models that are not parsimonious, ($ii$) challenges associated with auto-correlation inherent in the data, as well as ($iii$) complex dependency (cross-correlation) between the covariates that may be hard to capture.
Multivariate Time Series Forecasting Prediction Intervals +1
no code implementations • 7 Oct 2021 • Thabang Mathonsi, Terence L. Van Zyl
We benchmark our approach against the oft-preferred well-established statistical models.