Using large models and time series forecasting for market reaction analysis
- 1School of Economics and Management, Harbin Petroleum Institute, Harbin, Heilongjiang, China
- 2Accounting college, Harbin University of Commerce, Harbin, Heilongjiang, China
Abstract
Forecasting financial market movements requires integrating numerical time-series data with semantic information from financial narratives that shape investor behavior. While deep learning has advanced time-series prediction, existing methods often exploit trivial price autocorrelation rather than capturing genuine predictive signals from multimodal sources. We propose the Semantic-Time Fusion Architecture (STFA), which integrates structured market data with GPT-4-supervised FinBERT semantic representations of financial news through three key innovations: delay-aware temporal alignment between text publication and market reaction, semantic-guided attention mechanisms emphasizing news-relevant historical patterns, and multi-scale temporal fusion with multi-objective training to balance accuracy and interpretability. We evaluate our model on a large-scale financial news and stock price dataset using GPT-4 sentiment annotation across stratified high-liquidity stocks with strictly chronological train-validation-test splits (2013–2020 for training, 2021 for validation, 2022–2023 for testing). Sentiment annotations were generated independently for each article without using any future market information, ensuring strict temporal separation and preventing information leakage. Experiments demonstrate that STFA significantly outperforms traditional time-series models, multimodal baselines, and random walk benchmarks on return prediction, where price autocorrelation vanishes, rather than relying on price-level forecasting. Statistical validation with bootstrap confidence intervals, effect size analysis, and Bonferroni-corrected significance tests confirms robust improvements across independent runs. Ablation studies reveal that GPT-4-supervised FinBERT semantic representations provide superior signals compared with lexicon-based sentiment, with each architectural component contributing significantly to performance. This work demonstrates how temporally aligned semantic understanding enhances both predictive accuracy and model interpretability in financial forecasting.
Related
- Informed forecasting: Leveraging auxiliary knowledge to boost large language model performance on time series forecasting
- Short-term rental market occupancy - daily time series for 2017-2022 on 500 markets worldwide
- Hierarchical Time Series Forecasting with Operator-Valued Kernels
- DBCM: Dual-Branch Calibration Module for Test-Time Adaptation in Time Series Forecasting
- Coherent Hierarchical Forecasting for Proportion and Discrete Time Series
BibTeX
@article{he2026using,
title = {{Using large models and time series forecasting for market reaction analysis}},
author = {Hong He and Xinhai Li},
journal = {PLOS One},
year = {2026},
month = sep,
doi = {10.1371/journal.pone.0356139},
url = {https://doi.org/10.1371/journal.pone.0356139},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Using large models and time series forecasting for market reaction analysis
AU - Hong He
AU - Xinhai Li
PY - 2026
DA - 2026/09/30
JO - PLOS One
DO - 10.1371/journal.pone.0356139
UR - https://doi.org/10.1371/journal.pone.0356139
AB - Forecasting financial market movements requires integrating numerical time-series data with semantic information from financial narratives that shape investor behavior. While deep learning has advanced time-series prediction, existing methods often exploit trivial price autocorrelation rather than capturing genuine predictive signals from multimodal sources. We propose the Semantic-Time Fusion Architecture (STFA), which integrates structured market data with GPT-4-supervised FinBERT semantic representations of financial news through three key innovations: delay-aware temporal alignment between text publication and market reaction, semantic-guided attention mechanisms emphasizing news-relevant historical patterns, and multi-scale temporal fusion with multi-objective training to balance accuracy and interpretability. We evaluate our model on a large-scale financial news and stock price dataset using GPT-4 sentiment annotation across stratified high-liquidity stocks with strictly chronological train-validation-test splits (2013–2020 for training, 2021 for validation, 2022–2023 for testing). Sentiment annotations were generated independently for each article without using any future market information, ensuring strict temporal separation and preventing information leakage. Experiments demonstrate that STFA significantly outperforms traditional time-series models, multimodal baselines, and random walk benchmarks on return prediction, where price autocorrelation vanishes, rather than relying on price-level forecasting. Statistical validation with bootstrap confidence intervals, effect size analysis, and Bonferroni-corrected significance tests confirms robust improvements across independent runs. Ablation studies reveal that GPT-4-supervised FinBERT semantic representations provide superior signals compared with lexicon-based sentiment, with each architectural component contributing significantly to performance. This work demonstrates how temporally aligned semantic understanding enhances both predictive accuracy and model interpretability in financial forecasting.
SN - 1932-6203
ER -