FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization

Junyan Tan, Yifan Li, Minghao Wang, Zihan Chen, Haoyu Zhang · 2026

Read FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization by Junyan Tan, Yifan Li, Minghao Wang, Zihan Chen, Haoyu Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-28933.txt.

Introduction

Tan, Li, Wang, Chen and Zhang present FinInvest-GTCN, a model that helps venture capital investors judge start-ups with numbers rather than recommendations. Venture decisions mix many data sources, unstable time series and high stakes with little data, and investors need predictions they can explain. The architecture combines a graph encoder for the investment ecosystem's network of links, a temporal module that fuses patterns at several time scales, and a causal decision head that outputs risk-adjusted predictions.

Each prediction comes with explanations of which factors drove it. A meta-causal strategy fine-tunes the model for new, data-poor sectors, keeping updates aligned with plausible causal structure. On private venture datasets, the model lowered its main risk-adjusted error from just over three to about two and a half. A simulated portfolio's total return rose by nearly a fifth. Ablations confirmed each component mattered.

Catalog record

CREATORJunyan Tan, Yifan Li, Minghao Wang, Zihan Chen, Haoyu Zhang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE70 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.28933

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

FinInvest-GTCN: Explainable Graph-Temporal-Causal Modeling for Risk-Aware Investment Decision Optimization Junyan Tan, Yifan Li, Minghao Wang, Zihan Chen, Haoyu Zhang Abstract Venture capital (VC) investment decisions face distinct challenges, such as multi-source heterogeneous data, non-stationary time series, and the demand for explainable predictions in high-stakes, low-data settings. To overcome these issues, we introduce FinInvest-GTCN , a Graph-Temporal-Causal Network that redefines the task from content recommendation to quantitative risk-return assessment. This architecture combines a relational graph encoder to capture the investment ecosystem’s topology 53 ; 58 , a multi-scale temporal fusion module to handle long-term dependencies and non-stationarity 59 ; 14 , and a causal decision head that generates risk-adjusted predictions with interpretable causal attributions 55 ; 60 . A core innovation is the Meta-Causal Adaptation (MCA) strategy, which facilitates robust fine-tuning for new, data-scarce sectors by aligning updates with causally-plausible structures derived from meta-pretraining. Comprehensive experiments on proprietary VC datasets show that FinInvest-GTCN delivers state-of-the-art results, markedly lowering the primary Risk-Adjusted Mean Squared Error (RA-MSE) to 2.51 from a baseline of 3.05 and boosting the cumulative return of a simulated portfolio by 18.7%. Ablation studies underscore the essential role of each component, while additional analyses confirm the model’s stability, interpretability, and enhanced adaptability. This work pioneers a data-driven, explainable framework for investment decision support. 1 Introduction Financial investment decision-making, particularly in venture capital (VC), constitutes a high-stakes sequential prediction problem characterized by multi-source heterogeneous data and complex inter-asset relationships 2 ; 44 ; 75 . Practitioners must analyze time-series data encompassing financial metrics, team dynamics, and market signals, while simultaneously accounting for the broader ecosystem topology—including competitive overlaps and shared investor networks—to forecast risk-adjusted returns for potential assets. Despite the abundance of historical data, accurately modeling these sequential, relational, and non-stationary patterns to support robust and explainable decisions remains a formidable challenge. 81 ; 84 ; 82 ; 83 ; 80 ; 56 ; 77 ; 86 Prior work in sequential recommendation and financial modeling exhibits critical limitations when applied to this domain. Traditional tree-based approaches rely heavily on manual feature engineering, which is labor-intensive and fails to capture dynamic temporal dependencies 69 ; 37 ; 30 . Standard neural sequence models, such as LSTMs or Transformers, process observation histories but neglect the rich relational structure among assets 18 ; 79 ; 64 ; 61 ; 11 ; 76 ; 13 ; 85 ; 89 ; 32 ; 12 . Moreover, existing methods often optimize for simplistic objectives such as click-through or conversion rates, which fundamentally misalign with the investment goal of maximizing risk-adjusted returns. These approaches also typically lack mechanisms for explainability and struggle to adapt to new sectors with scarce data, thereby hindering practical deployment in evolving markets. Figure 1 : Paradigm shift from traditional investment decision methods to the proposed FinInvest-GTCN framework. Left: Existing methods suffer from fragmented approaches—manual feature engineering, isolated temporal modeling that ignores relational context, and lack of explainability. Right: Our unified framework integrates graph-based ecosystem topology modeling, multi-scale temporal fusion, and causal attribution, with Meta-Causal Adaptation (MCA) for robust fine-tuning, directly aligning with the goal of maximizing risk-adjusted returns. To address these gaps, we propose FinInvest-GTCN , a novel G raph- T emporal- C ausal N etwork for quantitative investment decision optimization. Our framework introduces three key innovations: (1) a relational graph encoder that captures the investment ecosystem’s topology via Graph Attention Networks; (2) a multi-scale temporal fusion module that learns patterns across different horizons using parallel Transformers; and (3) a causal decision head that predicts risk-adjusted returns while enabling approximate causal attribution for model explanations. Furthermore, we introduce a M eta- C ausal A daptation (MCA) strategy that regularizes fine-tuning towards causally-plausible structures, thereby enhancing robustness in low-data scenarios. Collectively, these contributions provide a principled, end-to-end solution that transcends traditional recommendation paradigms to directly address the core challenges of investment analysis. Extensive experiments on proprietary and simulated venture capital datasets demonstrate the effectiveness of our approach. FinInvest-GTCN achieves a state-of-the-art Risk-Adjusted MSE of 2.51, substantially outperforming strong baselines including repurposed sequential recommenders and financial factor models. Ablation studies confirm the importance of each architectural component, and our MCA strategy reduces RA-MSE by 16% compared to standard fine-tuning on a new, data-scarce sector. In a simulated online A/B test, a portfolio constructed using our model’s rankings yields an 18.7% cumulative return with lower volatility, underscoring its practical utility for real-world deployment. The remainder of this paper is organized as follows. Section 2 reviews related work. Section 3 details the FinInvest-GTCN architecture. Section 4 presents the experimental setup and results. Section 5 provides comprehensive ablation studies. Section 6 offers supplementary experiments, and Section 7 concludes the paper. 2 Related Work Our work lies at the intersection of three major research areas: graph-based financial modeling, temporal sequence learning for investment analysis, and causal inference for explainable predictions. We review each area and position our contributions accordingly. 2.1 Graph Neural Networks in Finance Graph neural networks (GNNs) have emerged as powerful tools for modeling relational structures in financial data 54 ; 43 ; 72 ; 74 ; 31 . Early applications focused on stock market prediction by constructing graphs based on industry sectors or correlation matrices. More recent works have extended GNNs to model complex inter-company relationships, including supply chains, competitive dynamics, and shared investor networks 47 ; 39 ; 38 ; 22 ; 6 . Graph Attention Networks (GATs) have proven particularly effective by learning adaptive edge weights that capture varying relationship strengths 66 ; 87 ; 70 ; 50 . However, existing approaches predominantly target public equity markets and short-term price prediction 62 ; 8 ; 17 , leaving the venture capital domain—characterized by sparse, irregular observations and longer investment horizons—largely unexplored. 2.2 Temporal Modeling for Financial Time Series Sequential modeling has long been central to financial forecasting. Traditional approaches relied on autoregressive models and handcrafted technical indicators. The advent of deep learning introduced LSTM and GRU architectures capable of capturing long-range dependencies in price sequences 16 ; 62 . More recently, Transformer-based models have demonstrated superior performance by leveraging self-attention mechanisms to model complex temporal patterns 14 ; 10 ; 40 ; 73 ; 68 . Multi-scale temporal modeling, which processes time series at different granularities simultaneously, has shown promise in capturing both short-term fluctuations and long-term trends 24 ; 21 ; 59 . Despite these advances, most methods focus on single-asset prediction and do not integrate the relational context essential for venture capital analysis, where an asset’s prospects depend heavily on its ecosystem position 3 ; 75 . 2.3 Causal Inference and Explainability in Machine Learning As machine learning models are increasingly deployed in high-stakes financial decisions, the demand for explainability has intensified. Post-hoc explanation methods such as SHAP and LIME provide feature attributions but lack causal grounding 60 ; 88 . Causal inference frameworks offer principled approaches to understanding model behavior through interventional reasoning 1 ; 41 ; 19 ; 9 ; 36 . Recent work has explored integrating causal structures into neural networks to improve both robustness and interpretability 49 ; 55 ; 48 . In the investment domain, explainability is particularly critical for regulatory compliance and building investor trust. However, existing financial models rarely incorporate causal reasoning, and methods for

Read the complete text