Probabilistic Time Series Forecasts with Autoregressive Transformation Models
This work addresses the need for reliable probabilistic forecasting in various applications, offering an incremental improvement by unifying existing research directions for better model interpretability.
The paper tackles the challenge of ensuring probabilistic time series forecasts are both expressive and interpretable by proposing Autoregressive Transformation Models (ATMs), which combine semi-parametric distribution assumptions with interpretable specifications, and demonstrates their properties through theoretical analysis and empirical evaluation on simulated and real-world datasets.
Probabilistic forecasting of time series is an important matter in many applications and research fields. In order to draw conclusions from a probabilistic forecast, we must ensure that the model class used to approximate the true forecasting distribution is expressive enough. Yet, characteristics of the model itself, such as its uncertainty or its feature-outcome relationship are not of lesser importance. This paper proposes Autoregressive Transformation Models (ATMs), a model class inspired by various research directions to unite expressive distributional forecasts using a semi-parametric distribution assumption with an interpretable model specification. We demonstrate the properties of ATMs both theoretically and through empirical evaluation on several simulated and real-world forecasting datasets.