---
title: Practical estimation methods for high-dimensional multivariate stochastic volatility models
url: https://www.ml-quant.com/papers/repec/cir-cirwor-2026s-20/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-11-14
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: RePEc:cir:cirwor:2026s-20
source_url: https://econpapers.repec.org/RePEc:cir:cirwor:2026s-20
featured: 2026-11-13
citations: unknown
topic: Econometrics & Forecasting
---


# Practical estimation methods for high-dimensional multivariate stochastic volatility models

Proposes computationally efficient estimators for multivariate stochastic volatility models using moment equations and shrinkage methods, avoiding numerical optimization while maintaining reliability.

- Source: https://econpapers.repec.org/RePEc:cir:cirwor:2026s-20
- Identifier: RePEc:cir:cirwor:2026s-20
- Released: 2026-11-02
- First featured: Quant Letter No. 139 (2026-11-13): https://www.ml-quant.com/issues/2026-11-13/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Econometrics & Forecasting
- Authors: MD Nazmul Ahsan, Jean-Marie Dufour

## Related

- [Online Multivariate Regularized Distributional Regression for High-dimensional Probabilistic Electricity Price Forecasting](https://www.ml-quant.com/papers/arxiv/2504.02518/): The study introduces a faster, more accurate online model for predicting electricity prices.
- [Forecasting with machine learning Shadow-Rate VARs](https://www.ml-quant.com/papers/ssrn/4828070/): The article examines the use of Shadow Rate Vector Autoregressions in macroeconomic forecasting, particularly the effects of shrinkage priors.
- [Regularization for electricity price forecasting](https://www.ml-quant.com/papers/arxiv/2404.03968/): The study reveals that LQ and elastic net penalty functions provide more precise electricity price predictions than other methods, including the popular LASSO. It also confirms that cross-validation is a useful tool for optimizing parameters.
- [Interpretive Earnings Forecasts via Machine Learning: A High-Dimensional Financial Statement Data Approach](https://www.ml-quant.com/papers/ssrn/4619313/): Using machine learning models and comprehensive Compustat financial statement data for earnings forecasting can yield predictions that are up to 13% more accurate than traditional linear approaches.
- [Bayesian precision factor models for high-dimensional macroeconomic time series](https://www.ml-quant.com/papers/arxiv/2611.12368/): VAPOR jointly estimates vector autoregressive dynamics and factor-structured precision to reveal direct dependencies in high-dimensional macroeconomic time series.
- [Behind the curve: Reassessing inflation risks with a High-dimensional Distributional VAR (HiDVAR)](https://www.ml-quant.com/papers/repec/zbw-imfswp-344237/): HiDVAR, a data-rich distributional framework, improves inflation tail-risk assessment by 28% in CRPS relative to standard benchmarks and performs well in real-time evaluation.