---
title: A Dynamic Regime-Switching Model Using Gated Recurrent Straight-Through Units
url: https://www.ml-quant.com/papers/ssrn/4810879/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 4810879
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4810879
featured: 2024-05-01
citations: 6
topic: Econometrics & Forecasting
---


# A Dynamic Regime-Switching Model Using Gated Recurrent Straight-Through Units

The Gated Recurrent Straightthrough Unit (GRSTU), a new deep learning model, outperforms statistical jump models in identifying regime changes in the S&P500 index, especially with smaller datasets.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4810879
- Identifier: SSRN 4810879
- Released: 2024-04-29
- First featured: Quant Letter No. 47 (2024-05-01): https://www.ml-quant.com/issues/2024-05-01/
- Citations (Semantic Scholar): 6
- Published in: not yet
- Topic: Econometrics & Forecasting

## Related

- [A Comparison of Standard Statistical, Machine Learning and Deep Learning Methods in Forecasting the Time Series](https://www.ml-quant.com/papers/ssrn/4840148/): The accuracy of Machine Learning and Deep Learning in forecasting macroeconomic indicators is compared to the traditional statistical method ARIMA.
- [Distributional Refinement Network: Distributional Forecasting via Deep Learning](https://www.ml-quant.com/papers/arxiv/2406.00998/): The article introduces the Distributional Refinement Network (DRN), a model that enhances predictive performance and interpretability in actuarial modelling by merging a baseline model with a flexible neural network.
- [Universality of Linear Recurrences Followed by Non-linear Projections: Finite-Width Guarantees and Benefits of Complex Eigenvalues](https://www.ml-quant.com/papers/arxiv/2307.11888/): The study reveals that deep neural networks using linear complex-valued RNNs and MLPs can accurately approximate regular causal sequence-to-sequence maps, with complex eigenvalues near unit disk aiding in information storage.
- [Downside risk reduction using regime-switching signals: a statistical jump model approach](https://www.ml-quant.com/papers/arxiv/2402.05272/): The article introduces a regime-switching investment strategy using a statistical jump model, which can reduce risk and increase returns compared to traditional strategies.
- [Quantum-Enhanced Forecasting: Leveraging Quantum Gramian Angular Field and CNNs for Stock Return Predictions](https://www.ml-quant.com/papers/arxiv/2310.07427/): Quantum-Enhanced Forecasting for Time Series: The study introduces a time series forecasting method, Quantum Gramian Angular Field (QGAF), that merges quantum computing and deep learning to enhance the accuracy of time series classification and forecasting, and validates its effectiveness using major stock market datasets.
- [Quantile deep learning models for multi-step ahead time series prediction](https://www.ml-quant.com/papers/arxiv/2411.15674/): The article introduces a new deep learning framework for predicting multi-step time series, which improves the performance of deep learning models. It has been effectively tested on Bitcoin and Ethereum, demonstrating its ability to manage volatility and provide useful information for decision-making.