FinBench: Time-Gated Calibration and Uncertainty Benchmarking for Agentic Financial Forecasting

• URL: http://arxiv.org/abs/2607.16229v1
• Date: Wed, 24 Jun 2026 01:27:30 GMT
• Title: FinBench: Time-Gated Calibration and Uncertainty Benchmarking for Agentic Financial Forecasting
• Abstract summary: FinBench is a benchmark designed to evaluate calibration and uncertainty quality for financial forecasting.<n>This paper describes the benchmark specification and reports a small pilot run as a sanity check of the pipeline.
• Score: 0.0
• License: http://creativecommons.org/publicdomain/zero/1.0/
• Abstract: Large language models (LLMs) are increasingly used as components of agentic systems that observe, plan, and act. In finance, even "assistive" systems become decision-relevant once their outputs are used to size trades or allocate risk. A key failure mode is the confidence--competence gap: a model that is only slightly better than chance but consistently overconfident will, under typical bet-sizing rules, generate negative long-run growth. Existing benchmarks emphasize semantic understanding or point accuracy, but do not directly test probabilistic calibration under the temporal constraints and non-stationarity that define real markets. We introduce FinBench, a benchmark designed to evaluate calibration and uncertainty quality for financial forecasting in a setting that is (i) strictly time-gated to avoid look-ahead bias and (ii) evaluated with strictly proper scoring rules that penalize hallucinated confidence. FinBench tasks require models to output (a) a probability of positive return and (b) an 80% prediction interval for realized log return; evaluation uses the Brier score and the Winkler interval score, along with skill scores against hard baselines. This paper describes the benchmark specification and reports a small pilot run (one trading day; three liquid tickers; 33 forecasts) as a sanity check of the pipeline. The pilot illustrates how calibration-sensitive metrics distinguish between "confident but fragile" behavior and uncertainty-aware forecasting.

Related papers

• Latent-Regime Bias Auditing for Volatility Forecasting [0.44333156307871585]
  This paper proposes a model-agnostic audit framework for evaluating whether volatility forecasts remain reliable across latent market regimes.<n>We learn time-series representations of market-state windows, cluster them into regimes using only training information, and compare aggregate forecast behavior with regime-conditional bias, tail-underprediction, and underprediction-sensitive economic losses.
  arXiv  Detail & Related papers  (2026-08-03T02:09:39Z)
• FinAbstain: Uncertainty-Calibrated Multimodal RAG for Selective Financial Forecasting [8.891247829087936]
  FinAbstain is a framework for uncertainty-calibrated multimodal retrieval-augmented generation with selective prediction.<n>A point-in-time retriever admits only information public at the forecast timestamp and supplies modality-specific evidence to fundamental, news, technical, risk, and verification agents.<n>A controller predicts bullish, bearish, or neutral outcomes only when uncertainty is below a validated threshold.
  arXiv  Detail & Related papers  (2026-07-27T07:05:22Z)
• Expectations vs. Realities: The Cost of MSE-Optimal Forecasting Under Conditional Uncertainty [1.2489632787815885]
  Multi-step time series forecasting (MSF) is commonly evaluated using pointwise error metrics such as mean squared error (MSE)<n>We show that this can be misleading under conditional uncertainty, where the conditional expectation becomes unrepresentative of typical realized values at longer horizons.<n>This establishes a fundamental, model-agnostic trade-off between point accuracy and marginal realism in MSF evaluation.
  arXiv  Detail & Related papers  (2026-06-03T01:50:32Z)
• Conformal Reliability: A New Evaluation Metric for Conditional Generation [62.761166941396844]
  Conditional generative models have recently remarkable success in various applications.<n>We propose a novel evaluation metric called reliability based on prediction set at a pre-specified confidence level.<n>We introduce Conformal ReLiability (CReL), a framework that can (i) construct a prediction set and (ii) accurately optimize the reliability score within the constructed prediction set.
  arXiv  Detail & Related papers  (2026-05-29T03:52:44Z)
• Quantifying the Risk-Return Tradeoff in Forecasting [0.0]
  Average forecast accuracy is not the same as forecast reliability.<n>I treat forecast loss differentials relative to a benchmark as a return series.<n>I then evaluate these returns using risk-adjusted performance measures from finance, including the Sharpe ratio, Sortino ratio, Omega ratio, and drawdown-based metrics.
  arXiv  Detail & Related papers  (2026-05-10T19:21:12Z)
• Bi-Level Chaotic Fusion Based Graph Convolutional Network for Stock Market Prediction Interval [0.0]
  Financial market forecasting is inherently uncertain, yet most deep learning approaches rely on single estimates without uncertainty.<n>The problem can be solved using prediction intervals, which allow obtaining an uncertainty and lower bound for the prediction.<n>Yet, the current methods tend to disregard relationships between assets or cannot simultaneously ensure good calibration sharpness.
  arXiv  Detail & Related papers  (2026-05-05T05:06:42Z)
• Statistical Inference for Score Decompositions [0.0]
  We introduce methods for score decompositions, which partition scoring functions for predictive assessment into three interpretable components.<n>We find that for survey inflation forecasts, discrimination abilities can differ significantly even when overall predictive ability does not.<n>By disentangling forecast accuracy from backtest performance, the method exposes critical shortcomings in current banking regulation.
  arXiv  Detail & Related papers  (2026-03-04T16:57:08Z)
• Predicting Mortgage Default with Machine Learning: AutoML, Class Imbalance, and Leakage Control [7.5947844798897535]
  We compare multiple machine learning approaches for mortgage default prediction using a real-world loan-level dataset.<n>We employ leakage-aware feature selection, a strict temporal split that constrains both origination and reporting periods.<n>Across multiple positive-to-negative ratios, performance remains stable, and an AutoML approach achieves the strongest AUROC among the models evaluated.
  arXiv  Detail & Related papers  (2026-01-27T15:25:04Z)
• Consistency Checks for Language Model Forecasters [54.62507816753479]
  We measure the performance of forecasters in terms of the consistency of their predictions on different logically-related questions.<n>We build an automated evaluation system that generates a set of base questions, instantiates consistency checks from these questions, elicits predictions of the forecaster, and measures the consistency of the predictions.
  arXiv  Detail & Related papers  (2024-12-24T16:51:35Z)
• Revisiting Confidence Estimation: Towards Reliable Failure Prediction [53.79160907725975]
  We find a general, widely existing but actually-neglected phenomenon that most confidence estimation methods are harmful for detecting misclassification errors. We propose to enlarge the confidence gap by finding flat minima, which yields state-of-the-art failure prediction performance.
  arXiv  Detail & Related papers  (2024-03-05T11:44:14Z)
• Score Matching-based Pseudolikelihood Estimation of Neural Marked Spatio-Temporal Point Process with Uncertainty Quantification [59.81904428056924]
  We introduce SMASH: a Score MAtching estimator for learning markedPs with uncertainty quantification. Specifically, our framework adopts a normalization-free objective by estimating the pseudolikelihood of markedPs through score-matching. The superior performance of our proposed framework is demonstrated through extensive experiments in both event prediction and uncertainty quantification.
  arXiv  Detail & Related papers  (2023-10-25T02:37:51Z)
• When Does Confidence-Based Cascade Deferral Suffice? [69.28314307469381]
  Cascades are a classical strategy to enable inference cost to vary adaptively across samples. A deferral rule determines whether to invoke the next classifier in the sequence, or to terminate prediction. Despite being oblivious to the structure of the cascade, confidence-based deferral often works remarkably well in practice.
  arXiv  Detail & Related papers  (2023-07-06T04:13:57Z)
• Beyond calibration: estimating the grouping loss of modern neural networks [68.8204255655161]
  Proper scoring rule theory shows that given the calibration loss, the missing piece to characterize individual errors is the grouping loss. We show that modern neural network architectures in vision and NLP exhibit grouping loss, notably in distribution shifts settings.
  arXiv  Detail & Related papers  (2022-10-28T07:04:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.