Conformalised Conditional Normalising Flows for Joint Prediction Regions in time series

TL;DR

This paper introduces a novel conformal prediction method for conditional normalising flows, enabling the construction of valid, potentially disjoint multi-step time series prediction regions with improved efficiency.

Contribution

It proposes a new approach to conformalise conditional normalising flows for time series, addressing the challenge of obtaining valid, multimodal prediction regions.

Findings

Generates disjoint, multimodal prediction regions

Achieves finite-sample validity guarantees

Improves predictive efficiency in time series forecasting

Abstract

Conformal Prediction offers a powerful framework for quantifying uncertainty in machine learning models, enabling the construction of prediction sets with finite-sample validity guarantees. While easily adaptable to non-probabilistic models, applying conformal prediction to probabilistic generative models, such as Normalising Flows is not straightforward. This work proposes a novel method to conformalise conditional normalising flows, specifically addressing the problem of obtaining prediction regions for multi-step time series forecasting. Our approach leverages the flexibility of normalising flows to generate potentially disjoint prediction regions, leading to improved predictive efficiency in the presence of potential multimodal predictive distributions.