# Dynamic Covariance Models for Multivariate Financial Time Series

arXiv: https://arxiv.org/abs/1305.4268 · HTML: https://mlsanity.com/paper/1305.4268/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 May 2013.

## Built on by (most cited first, 9 of 9)

- [A Survey on Time-Series Pre-Trained Models](https://mlsanity.com/paper/2305.10716/) (2023)
- [Text2TimeSeries: Enhancing Financial Forecasting through Time Series Prediction Updates with Event-Driven Insights from Large Language Models](https://mlsanity.com/paper/2407.03689/) (2024)
- [Quantile-based fuzzy C-means clustering of multivariate time series: Robust techniques](https://mlsanity.com/paper/2109.11027/) (2021)
- [DTAAD: Dual Tcn-Attention Networks for Anomaly Detection in Multivariate Time Series Data](https://mlsanity.com/paper/2302.10753/) (2023)
- [Mitigating Cold-start Forecasting using Cold Causal Demand Forecasting Model](https://mlsanity.com/paper/2306.09261/) (2023)
- [Neural Generalised AutoRegressive Conditional Heteroskedasticity](https://mlsanity.com/paper/2202.11285/) (2022)
- [No Imputation Needed: A Switch Approach to Irregularly Sampled Time Series](https://mlsanity.com/paper/2309.08698/) (2023)
- [Variational Heteroscedastic Volatility Model](https://mlsanity.com/paper/2204.05806/) (2022)
- [DeLELSTM: Decomposition-based Linear Explainable LSTM to Capture Instantaneous and Long-term Effects in Time Series](https://mlsanity.com/paper/2308.13797/) (2023)

## Abstract (arXiv)

The accurate prediction of time-changing covariances is an important problem in the modeling of multivariate financial data. However, some of the most popular models suffer from a) overfitting problems and multiple local optima, b) failure to capture shifts in market conditions and c) large computational costs. To address these problems we introduce a novel dynamic model for time-changing covariances. Over-fitting and local optima are avoided by following a Bayesian approach instead of computing point estimates. Changes in market conditions are captured by assuming a diffusion process in parameter values, and finally computationally efficient and scalable inference is performed using particle filters. Experiments with financial data show excellent performance of the proposed method with respect to current standard models.