From Vector Autoregressions to AI-based Time Series Forecasting: A Review

2026-07-15

AI Summary

This study addresses the challenges of modeling high-dimensional, nonstationary, and nonlinear time series by systematically comparing traditional vector autoregressive (VAR) models with three cutting-edge AI approaches—Transformers, zero-shot pretrained large models, and diffusion generative models—within a unified conditional distribution framework. Through comparative analysis of their capabilities in dynamic modeling, information utilization, and predictive distribution representation, the work elucidates the performance advantages of AI methods while highlighting their limitations in structural inference and policy analysis. The research underscores the irreplaceable role of econometric models in interpretability, hypothesis testing, and policy simulation, and proposes key directions for integrating AI and econometric paradigms to harness the strengths of both.

Abstract

Forecasting is a central goal of time-series analysis. This review centers on three major developments in recent AI-based time-series forecasting: transformers, large pretrained models for zero-shot forecasting, and diffusion-based generative forecasters. We connect these methods to the econometric tradition built around the vector autoregression (VAR) through a common object: the conditional distribution of the future given the past. The review is organized around three long-standing challenges: \emph{high dimensionality}, \emph{nonstationarity}, and \emph{nonlinearity}. We argue that modern methods make progress by expanding the classical forecasting template: they allow more flexible dynamics, use larger information sets and training corpora, and represent richer predictive distributions. Yet they often lack the inferential and structural tools that make classical models useful for testing, explanation, and policy analysis. We close by outlining open problems where econometric tools remain important.

Problem

• high dimensionality
• nonstationarity
• nonlinearity

Innovation

• transformers
• large pretrained models
• diffusion-based forecasting
• vector autoregression
• conditional distribution