Shocks that spread through the web, such as cybersecurity breach disclosures, can abruptly disrupt financial time series and cause substantial abnormal losses. While these events are disclosed as discrete records through news reports, regulatory filings, or public databases, their consequences unfold through continuous market dynamics. This creates an event-conditioned impact prediction problem: given pre-event market history and limited event metadata, the goal is to estimate short-term post-disclosure abnormal loss rather than reconstruct the full post-event trajectory. However, most time-series forecasting models focus on endogenous regularities such as trend, seasonality, and autocorrelation, and thus struggle with rare and heterogeneous external events. The challenge is further amplified by sparse high-impact events and background market noise. We introduce EventTime, a multi-resolution framework that combines long-horizon market context, short-horizon pre-event dynamics, and event metadata. It incorporates an event fusion module that couples temporal representations with event attributes to identify relevant recent market patterns. To mitigate sparse supervision, EventTime further introduces a dynamic contrastive objective that constructs event- and time-series-aware positive and negative pairs during training. We also construct SECURE, a real-world dataset aligning cybersecurity incidents with stock-market time series and structured and LLM-derived semantic features. Experiments show that EventTime consistently outperforms state-of-the-art time-series and event-aware baselines in estimating post-event financial losses. Further analyses demonstrate more event-sensitive representations, greater robustness to incomplete metadata, and more interpretable estimates of short-term market impact following cybersecurity disclosures.
Figures & tables
Figure 1. Stock market shock induced by a cybersecurity event: Facebook’s 2018 data breach.
Figure 2. Overall framework of EventTime.
Figure 3. Analysis of EventTime model performance and robustness.
Model
R2
MAE
RMSE
IC
Rank IC
DLinear
-3.5240
0.0315
0.0395
0.1285
0.0716
TimeMixer
0.0154
0.0125
0.0160
0.1850
0.2570
WPMixer
0.1125
0.0121
0.0152
0.3910
0.3855
Autoformer
0.1450
0.0122
0.0149
0.4310
0.2185
PatchTST
0.2150
0.0120
0.0143
0.3655
0.2980
iTransformer
0.2280
0.0118
0.0142
0.4320
0.3750
Table 1. Comparison of Backbones Under Identical Finetuning Strategy.
Training
R2
MAE
RMSE
IC
Rank IC
EventTime w/o e
0.2455
0.0115
0.0145
0.4850
0.4950
EventTime(s)
0.3212
0.0117
0.0135
0.5145
0.4999
EventTime(-)
0.4152
0.0102
0.0128
0.6580
0.5525
EventTime
0.5242
0.0096
0.0119
0.7446
0.5579
Table 2. Comparison of Training Strategies for EventTime. We denote EventTime(s) as the variant trained with supervised loss only, EventTime(-) as the variant using heuristic contrastive similarity, and EventTime as the full model.
Figure 4. Post-disclosure excess-return dynamics. The target impact label is defined as the maximum excess-return decline.
Figure 5. Representative stock-price trajectories and predicted short-term losses around cybersecurity disclosures.
Model
R2
MAE
RMSE
DLinear
0.2736
10.9538
18.1245
TimeMixer
0.2982
10.2326
17.8543
WPMixer
0.2822
10.4819
18.0169
Autoformer
0.2941
9.5585
17.8676
PatchTST
0.3070
10.1851
17.7033
iTransformer
0.3078
10.5662
17.6930
Table 3. Cross-domain evaluation on CAMELS for hydrological event impact prediction.
Accurately forecasting the impact of salient financial events on markets is critical for investors and policymakers. However, existing multimodal time-series models typically fuse text and prices symmetrically, without an explicit way to decide when event text is truly predictive, and thus struggle to exploit the directional event-to-price structure and the heterogeneous roles of textual and price signals. In this work, we propose GS-Fuse, a multimodal event-based forecasting framework that employs (i) a Granger-supervised, causal-aware gated fusion module, which learns to open toward event text only when it provides incremental predictive value beyond historical prices, and (ii) a multi-granularity alignment mechanism that jointly aligns high-level event representations and fine-grained textual cues with future market trajectories. Built as a flexible, plug-and-play adapter on top of off-the-shelf large language models and time-series foundation models, GS-Fuse can be instantiated across diverse backbones and market settings. Extensive experiments on real-world financial datasets show that GS-Fuse consistently outperforms state-of-the-art time-series and multimodal baselines across multiple assets and forecasting horizons.
Yang Zhang, En Chun, Ziyun Mao +2
Southwestern University of Finance and Economics Chengdu, China
Real-world time series are frequently driven by exogenous events and structural shifts, rendering conventional forecasting based solely on historical numerical observations insufficient. While language models can retrieve external news, standard retrieval-augmented approaches struggle with high noise, missing signals, and an inability to reason causally about event impacts. We propose SEER (Self-Evolving Event Reasoning and Retrieval), a closed-loop framework that dynamically optimizes event conditioning for time series forecasting. SEER translates prediction errors into two decoupled feedback mechanisms: (i) a reflective retrieval memory that refines subsequent search queries and filters spurious noise, and (ii) a persistent causal knowledge base that distills transferable domain dynamics. SEER enforces strict chronological boundaries across both event retrieval and reflection, preventing look-ahead bias and data leakage. Across six volatile time-series benchmarks, SEER consistently outperforms state-of-the-art time series foundation models and language model baselines.
Financial time series evolve across multiple temporal resolutions, challenging forecasting systems to incorporate newly available information without repeatedly recomputing unchanged representations. We introduce HARN, a Hierarchical Associative Resonance Network for event-driven multi-timeframe forecasting. HARN maintains persistent representations across temporal levels and updates each level only when its corresponding completed bar becomes available. The architecture combines causal multi-scale temporal encoding, gated associative memory, cross-level resonance, and hierarchical evidence aggregation, with forecasting performed in basis-point space and reconstructed to the original price scale. We evaluate HARN on four assets spanning equity, foreign exchange, and commodity markets using multiple random seeds and component ablations. HARN achieves competitive reconstructed-price forecasting errors against single-timeframe PatchTST and TimeXer baselines, while ablations reveal the effects of removing individual components across assets and timeframes. A code-level audit further examines consistency between the implementation and the defined event-driven causal protocol. The results position HARN as a persistent multi-timeframe forecasting framework rather than evidence of universal predictive superiority.