GLoR-forecast: state-contingent global-local text routing for financial and socio-economic time-series forecasting
- 1School of Economics and Management, China University of Geosciences
Abstract
Financial and socio-economic forecasting often requires information beyond numerical histories, as policy statements, news reports, institutional releases, and event descriptions may signal structural regimes or short-term shocks. Recent multimodal time-series methods incorporate timestamp-aligned texts as auxiliary variables in numerical forecasting backbones, but typically treat all textual signals as homogeneous local covariates. This overlooks an important distinction between local event-level text and broader contextual narratives. This paper proposes GLoR-Forecast (Global-Local Text Routing for Forecasting), a state-contingent textual fusion framework for multimodal time-series forecasting. The model encodes timestamp-aligned texts with a frozen language model, projects them into compact local textual variables, and attentively aggregates them into a global contextual representation. A text router then uses the current numerical time-series state as a query to dynamically weight local and global textual features. The fused representation augments the numerical sequence and can be used with standard forecasting backbones; we instantiate it with iTransformer. Experiments on nine Time-MMD domains show consistent gains over TaTS, reducing average MSE by 5.8% and MAE by 4.7%. The strongest improvements appear in Economy, Social Good, Traffic, and Agriculture, suggesting that state-dependent textual routing helps exploit both slow-moving structural context and short-term event shocks.
Related
- Forecasting European temperature-related mortality in Summer 2024: data-driven vs. physics-based forecast approaches
- Financial toxicity and the economic burden of Chagas disease: a scoping review.
- Graph-augmented transformer networks and explainable AI for economic impact forecasting in disrupted supply chains
- Navigating fragmented and contingent support: A qualitative study of British Armed Forces families raising neurodivergent children
- Shape-preserving minimum trace (SP-MinT): a regularized forecast reconciliation method for hierarchical time series
BibTeX
@article{li2026glorforecast,
title = {{GLoR-forecast: state-contingent global-local text routing for financial and socio-economic time-series forecasting}},
author = {Jianing Li},
journal = {Scientific Reports},
year = {2026},
month = sep,
doi = {10.1038/s41598-026-67974-0},
url = {https://doi.org/10.1038/s41598-026-67974-0},
note = {CC BY},
}
RIS
TY - JOUR
TI - GLoR-forecast: state-contingent global-local text routing for financial and socio-economic time-series forecasting
AU - Jianing Li
PY - 2026
DA - 2026/09/01
JO - Scientific Reports
DO - 10.1038/s41598-026-67974-0
UR - https://doi.org/10.1038/s41598-026-67974-0
AB - Financial and socio-economic forecasting often requires information beyond numerical histories, as policy statements, news reports, institutional releases, and event descriptions may signal structural regimes or short-term shocks. Recent multimodal time-series methods incorporate timestamp-aligned texts as auxiliary variables in numerical forecasting backbones, but typically treat all textual signals as homogeneous local covariates. This overlooks an important distinction between local event-level text and broader contextual narratives. This paper proposes GLoR-Forecast (Global-Local Text Routing for Forecasting), a state-contingent textual fusion framework for multimodal time-series forecasting. The model encodes timestamp-aligned texts with a frozen language model, projects them into compact local textual variables, and attentively aggregates them into a global contextual representation. A text router then uses the current numerical time-series state as a query to dynamically weight local and global textual features. The fused representation augments the numerical sequence and can be used with standard forecasting backbones; we instantiate it with iTransformer. Experiments on nine Time-MMD domains show consistent gains over TaTS, reducing average MSE by 5.8% and MAE by 4.7%. The strongest improvements appear in Economy, Social Good, Traffic, and Agriculture, suggesting that state-dependent textual routing helps exploit both slow-moving structural context and short-term event shocks.
SN - 2045-2322
ER -