---
title: Time-Varying Quantile VAR Models: Estimation and Inference
url: https://www.ml-quant.com/papers/arxiv/2610.36634/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-10-30
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2610.36634
source_url: https://arxiv.org/abs/2610.36634
featured: 2026-10-30
citations: unknown
topic: Econometrics & Forecasting
---


# Time-Varying Quantile VAR Models: Estimation and Inference

The research develops impulse response analysis for quantile VARs and finds that downside asymmetry in credit spreads and industrial production diminishes at longer forecast horizons.

- Source: https://arxiv.org/abs/2610.36634
- Identifier: arXiv:2610.36634
- Released: 2026-10-28
- First featured: Quant Letter No. 137 (2026-10-30): https://www.ml-quant.com/issues/2026-10-30/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Econometrics & Forecasting
- Authors: Jiti Gao, Fei Liu, Bin Peng

## Abstract (arXiv, CC0)

We study time-varying quantile VAR models in which the conditional quantile does not propagate linearly through the recursion, and develop a toolkit for impulse response analysis, which to the best of our knowledge has not been established elsewhere. For example, we establish Bahadur representation for quantile VAR regression, infer model parameters via smoothed check function, and define and investigate the best linear projection of the time-varying tau-quantile impulse response via local projections. An application to U.S. industrial production growth and credit spreads from 1919-2025 uncovers structural shifts in tail relationships, demonstrating downside asymmetry diminishes at longer forecast horizons.

## Related

- [TKAN: Temporal Kolmogorov-Arnold Networks](https://www.ml-quant.com/papers/ssrn/4825654/): The article presents Temporal Kolomogorov-Arnold Networks (TKANs), a new neural network design that merges the benefits of Recurrent Neural Networks and Long Short-Term Memory for improved multistep time series forecasting.
- [Deep learning models for price forecasting of financial time series: A review of recent advancements: 2020–2022](https://www.ml-quant.com/papers/arxiv/2305.04811/): The article explores deep learning models for financial forecasting, discussing their structures, uses, pros and cons, and potential future research areas.
- [Robust agents learn causal world models](https://www.ml-quant.com/papers/arxiv/2402.10877/): The research suggests that intelligent agents must learn an approximate causal model to generalize to new domains, impacting fields like transfer learning and causal inference.
- [Directional-shift Dirichlet ARMA models for compositional time series with structural break intervention](https://www.ml-quant.com/papers/arxiv/2601.16821/): The article introduces a new Bayesian model that analyzes compositional time series data, effectively handling structural breaks and enhancing forecasting accuracy during these changes.
- [Blameocracy: Causal Rhetoric in Politics](https://www.ml-quant.com/papers/arxiv/2504.06550/): U.S. Causal Rhetoric: Growing blame/credit language in congressional tweets changed donation patterns, fueled protests and polarization, and shifted public trust.
- [FinCARE: Financial Causal Analysis with Reasoning and Evidence](https://www.ml-quant.com/papers/arxiv/2510.20221/): KG+LLM for Financial Causal Discovery: Combines SEC knowledge graphs, LLM reasoning, and causal discovery to build more accurate finance‑grounded causal models.