---
title: High-Dimensional Approximate Factor Models with Score-Driven Dynamics: Estimation and Inference
url: https://www.ml-quant.com/papers/arxiv/2611.20030/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-11-20
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2611.20030
source_url: https://arxiv.org/abs/2611.20030
featured: 2026-11-20
citations: unknown
topic: Econometrics & Forecasting
---


# High-Dimensional Approximate Factor Models with Score-Driven Dynamics: Estimation and Inference

The study combines principal components extraction with score-driven model estimation to capture nonlinearities and heavy tails, delivering valid asymptotic prediction intervals for high-dimensional time series.

- Source: https://arxiv.org/abs/2611.20030
- Identifier: arXiv:2611.20030
- Released: 2026-11-17
- First featured: Quant Letter No. 140 (2026-11-20): https://www.ml-quant.com/issues/2026-11-20/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Econometrics & Forecasting
- Authors: Matteo Barigozzi, Enzo D'Innocenzo

## Abstract (arXiv, CC0)

We propose a factor model for high-dimensional time series where the dynamics of the latent factors is generated by a multivariate score-driven model, thus allowing to model nonlinearities and heavy tails in both the conditional mean and the conditional variance. Estimation is in two steps: first the factors are extracted via Principal Components and then the parameters of the score-driven model are estimated via Quasi Maximum Likelihood based on the estimated factors. Since factors are not identified, we first establish set-consistency and asymptotic normality for the estimated parameters governing their dynamics, as both the number of time series and the sample size diverge to infinity. Then, we construct valid asymptotic prediction intervals, which do not depend on the chosen identifying constraints for the factors. Numerical results confirm the goodness of our estimator.

## Related

- [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.
- [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.
- [Practical estimation methods for high-dimensional multivariate stochastic volatility models](https://www.ml-quant.com/papers/repec/cir-cirwor-2026s-20/): Proposes computationally efficient estimators for multivariate stochastic volatility models using moment equations and shrinkage methods, avoiding numerical optimization while maintaining reliability.
