← Back to all news
arXiv Statistics ML By Rohan Hemant Chhatre, Chiranjit Dutta, Nalini Ravishanker, Sumanta Basu

GPU-accelerated Regularized Vector Multiplicative Error Models for Positive-valued Financial Time Series

Read the original on arXiv Statistics ML →

The paper introduces a GPU‑accelerated regularized estimation framework for the logarithmic vector multiplicative error model (log‑vMEM) with multivariate Gamma errors, designed to handle high‑dimensional positive‑valued financial time series. It proposes three hierarchical lag structures—componentwise, own‑other, and elementwise—combined with group lasso, group MCP, group SCAD, and adaptive group lasso penalties to induce sparsity and capture heterogeneous lag dynamics. Parameter estimation uses an alternating block‑coordinate algorithm with sequential Gauss‑Seidel updates, while GPU‑accelerated numerical quadrature mitigates the computational load of evaluating the multivariate Gamma likelihood; simulation studies assess parameter recovery and lag selection, and a real‑data application on Microsoft’s realized volatility demonstrates the approach in practice.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Statistics ML.

Related stories

arXiv Statistics ML
Oct 7

Scalable Regularized Vector Multiplicative Error Models for Positive-valued Financial Time Series

The paper introduces scalable regularized estimation techniques for logarithmic multiplicative error models (log‑vMEM) applied to multivariate positive‑valued financial time series. It employs hierarchical lag structures and various group penalties, using a blockwise coordinate descent algorithm with Gauss‑Seidel updates to efficiently estimate parameters. GPU‑accelerated quadrature integration is used to overcome the main computational bottleneck, and the methods are evaluated through extensive simulations and applied to intraday realized volatility data for Microsoft.

By Rohan Hemant Chhatre, Chiranjit Dutta, Nalini Ravishanker, Sumanta Basu
agents
More like this →
arXiv Statistics ML
Aug 31

Autotune: fast, accurate, and automatic tuning parameter selection for Lasso

The paper introduces μs’ autotune, an automatic tuning strategy for the Lasso that optimizes a penalized Gaussian log‑likelihood over regression coefficients and noise standard deviation. Extensive simulations on regression and VAR models show that autotune is faster and yields better generalization and model selection, especially in low signal‑to‑noise regimes. The method also delivers a new noise‑standard‑deviation estimator, a visual diagnostic for sparsity, and is demonstrated on a real‑world financial dataset, with an accompanying R package available on GitHub.

By Tathagata Sadhukhan, Ines Wilms, Stephan Smeekes, Sumanta Basu
efficiency
More like this →
arXiv Machine Learning
Aug 7

Scalable estimation of VARMA models

arXiv:2608. 06340v1 Announce Type: cross Abstract: Vector autoregressive moving-average (VARMA) models have long been considered impractical beyond moderate dimensions: the likelihood is non-convex, the parametrization is identified only up to equivalence, and every evaluation costs a pass over the entire series.

By Daniel Paulin, Victor Elvira
safety
More like this →
Hugging Face Trending Papers
Aug 6

Scalable estimation of VARMA models

Vector autoregressive moving-average (VARMA) models have long been considered impractical beyond moderate dimensions: the likelihood is non-convex, the parametrization is identified only up to equivalence, and every evaluation costs a pass over the entire series. Yet their moving-average term captures with a few parameters what a pure autoregression matches only with many lags.

safety
More like this →