6 papers
SPACE: Sample-cloud Predictive Adaptive Conformal Ellipsoids for Multivariate Time-Series Forecasting
Baishi Li, Kelvin J. L. Koa, Ke-Wei Huang
Modern probabilistic time-series forecasters often express uncertainty through forecast samples. While typically converted into nominal prediction regions using empirical quantiles…
The Proxy Presumption: From Semantic Embeddings to Valid Social Measures
Baishi Li, Ta Yu, Kelvin J. L. Koa +1
Natural Language Processing is rapidly evolving into a primary instrument for Computational Social Science, with researchers increasingly using embeddings to measure latent constru…
FinStressTS: A Parametric Synthetic Benchmark for Time-Series Forecasting in Finance
Jiaze Sun, Kelvin J. L. Koa, Ruiyang Ni +3
Financial forecasting is difficult due to low signal-to-noise ratios, latent factors, heavy tails, regime shifts, and jumps. Real-world benchmarks offer limited failure attribution…
Reasoning on Time-Series for Financial Technical Analysis
Kelvin J. L. Koa, Jan Chen, Yunshan Ma +2
While Large Language Models have been used to produce interpretable stock forecasts, they mainly focus on analyzing textual reports but not historical price data, also known as Tec…
FinDeepForecast: A Live Multi-Agent System for Benchmarking Deep Research Agents in Financial Forecasting
Xiangyu Li, Xuan Yao, Guohao Qi +16
Deep Research (DR) Agents powered by advanced Large Language Models (LLMs) have fundamentally shifted the paradigm for completing complex research tasks. Yet, a comprehensive and l…
Temporal Relational Reasoning of Large Language Models for Detecting Stock Portfolio Crashes
Kelvin J. L. Koa, Yunshan Ma, Yi Xu +3
Stock portfolios are often exposed to rare consequential events (e.g., 2007 global financial crisis, 2020 COVID-19 stock market crash), as they do not have enough historical inform…