Stanza: A Nonlinear State Space Model for Probabilistic Inference in Non-Stationary Time Series
This work addresses the problem of probabilistic inference and forecasting in complex time series for domains requiring interpretability, though it appears incremental as an intermediate approach between traditional and deep learning methods.
The authors tackled the challenge of modeling long-term structure in non-stationary time series by proposing Stanza, a nonlinear state space model that balances forecasting accuracy with probabilistic inference. Stanza achieved competitive forecasting accuracy compared to deep LSTMs on real-world datasets, particularly for multi-step ahead forecasting.
Time series with long-term structure arise in a variety of contexts and capturing this temporal structure is a critical challenge in time series analysis for both inference and forecasting settings. Traditionally, state space models have been successful in providing uncertainty estimates of trajectories in the latent space. More recently, deep learning, attention-based approaches have achieved state of the art performance for sequence modeling, though often require large amounts of data and parameters to do so. We propose Stanza, a nonlinear, non-stationary state space model as an intermediate approach to fill the gap between traditional models and modern deep learning approaches for complex time series. Stanza strikes a balance between competitive forecasting accuracy and probabilistic, interpretable inference for highly structured time series. In particular, Stanza achieves forecasting accuracy competitive with deep LSTMs on real-world datasets, especially for multi-step ahead forecasting.