---
title: A Semiparametric Real-Time GARCH Model with Nonstationary Variance
url: https://www.ml-quant.com/papers/arxiv/2610.36739/
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:2610.36739
source_url: https://arxiv.org/abs/2610.36739
featured: 2026-10-30
citations: unknown
topic: Econometrics & Forecasting
---


# A Semiparametric Real-Time GARCH Model with Nonstationary Variance

A two-step estimator combines kernel smoothing with GARCH to capture gradual variance changes while allowing short-run volatility to respond to current information.

- Source: https://arxiv.org/abs/2610.36739
- Identifier: arXiv:2610.36739
- 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: Muyi Li, Yajia Chen, Yuqing Xu, Feiyu Jiang

## Abstract (arXiv, CC0)

We propose a semiparametric real-time GARCH (SRT-GARCH) model that combines a smooth deterministic long-run volatility component with a stationary short-run process driven by contemporaneous return innovations. The model accommodates gradual changes in unconditional variance while allowing short-run volatility to respond directly to current information rather than relying solely on lagged observations. We derive the variance normalization, conditional moments, and closed-form conditional density of the model. For estimation, we develop a two-step procedure that combines kernel smoothing with conditional Gaussian maximum likelihood. Under Gaussian innovations and mild regularity conditions, we establish consistency and asymptotic normality of the short-run parameter estimator. The analysis exploits geometric moment contraction to control serial dependence and shows that estimating the long-run component adds a positive semidefinite rank-one adjustment to the oracle covariance matrix obtained when the long-run component is known. The resulting asymptotic covariance matrix admits a plug-in estimator that requires neither lag truncation nor additional tuning parameters. Monte Carlo simulations show good finite-sample performance of the estimator and the associated inference procedure. Applications to FTSE~100 returns and U.S. natural gas futures show that allowing for a time-varying long-run component improves forecasting relative to the parametric RT-GARCH benchmark, while the additional real-time feedback mechanism provides more pronounced gains in the natural gas application than in the FTSE application. The proposed model also delivers competitive one-step-ahead Value-at-Risk forecasts.

## Related

- [Deep Learning Enhanced Multivariate GARCH](https://www.ml-quant.com/papers/arxiv/2506.02796/): A new volatility modeling framework, LSTM-BEKK, is introduced, integrating deep learning into multivariate GARCH processes for improved robustness and forecasting in financial return data.
- [Data-Driven Risk Measurement by SV-GARCH-EVT Model](https://www.ml-quant.com/papers/arxiv/2201.09434/): A new financial risk measurement model that considers fat-tailed distribution and leverage effect performs better than other models in capturing financial return characteristics.
- [StaFIR: Convex Learning of Stationarity-Aware Causal Filters](https://www.ml-quant.com/papers/arxiv/2610.07430/): A convex method learns causal filters that balance stationarity and input preservation for financial time series, adapting filtering strength to persistence while preserving signal similarity.
- [Price Conflict and US Stock Return Volatility Forecasting: Insights from over 150 Years with a Mixed-Frequency Framework](https://www.ml-quant.com/papers/repec/pre-wpaper-202620/): The GARCH-MIDAS model incorporating a quarterly news-based Price Conflict Index outperforms benchmarks for forecasting US stock volatility over 150 years of monthly and daily data.
- [Network Realized GARCH--Itô Models: Volatility Spillovers with High-Frequency Identification](https://www.ml-quant.com/papers/arxiv/2609.29515/): Introduces a network realized GARCH-Itô model that identifies dynamic volatility transmission among assets using high-frequency data, outperforming recursive forecasts on sector ETFs.
- [Multivariate Affine GARCH](https://www.ml-quant.com/papers/ssrn/5260415/): A specific financial model can capture time-varying volatility and dynamic correlation across asset returns, useful for portfolio optimization and option pricing.