An end-to-end data engineering pipeline and multivariate time-series forecasting engine utilizing PyTorch Forecasting and Temporal Fusion Transformers for high-scale demand modeling.

Abstract: Multivariate time series forecasting estimates future development by capturing variable relationships and constructing temporal regular, which is widely used in many scenarios, including ...

Abstract: Probabilistic forecasting of multivariate time series is essential for various downstream tasks. Most existing approaches rely on the sequences being uniformly spaced and aligned across all ...

When cross-asset correlations break down, which assets decouple first and which follow — and can we identify systemic stress regimes in real time? An end-to-end data engineering project that ingests ...