用于冻结时间序列基础模型的混合神经-经典校正：高频股票预测的全面消融研究
Hybrid Neural-Classical Correction for Frozen Time Series Foundation Models: A Comprehensive Ablation Study on High-Frequency Stock Prediction
AI总结：
本研究针对高频股票预测任务，对冻结的TimesFM模型开展混合神经-经典校正的全面消融实验，发现经典残差学习贡献最大，GatedLinear+RF性能最优，为基础模型适配提供了实用指导。
AI中文摘要：
时间序列预测的基础模型展现出出色的零样本泛化能力，但在高频金融等专业领域往往表现不佳。我们开展了一项全面研究，探索混合神经-经典校正方法，以将冻结的TimesFM（2亿参数）适配至波动开盘交易时段的股票收益预测任务。我们对比了两种神经校正架构——AttnCorrect（多头自注意力，约47.1万参数）与GatedLinear（带门控的低秩双线性投影，约4.9万参数），每种架构均辅以Random Forest残差学习。通过对10只主要科技股（NVDA、MSFT、AAPL、GOOG、GOOGL、AMZN、META、AVGO、TSLA、NFLX）的系统消融实验，覆盖200万个数据点，我们揭示了关键见解：（1）混合神经-经典方法实现了0.597的池化相关性，较冻结的TimesFM提升了6.4倍的日均相关性；（2）经典残差学习（Random Forest）是贡献最大的单一组件，其效果可与神经校正组件相当或超过；（3）当移除经典残差学习时，更简单的神经架构出人意料地优于复杂架构；（4）自注意力是纯神经贡献中最大的部分。GatedLinear+RF实现了最佳整体性能，其神经参数比AttnCorrect+RF少9倍。我们报告了三种互补的相关性指标——日均、跨日累积和池化，以全面呈现预测质量。我们的研究提供了实用指导：有效的基础模型适配需要谨慎整合神经与经典组件，其中经典方法发挥着关键的互补作用。
英文摘要：
Foundation models for time series forecasting demonstrate impressive zero-shot generalization but often underperform on specialized domains such as high-frequency finance. We present a comprehensive study of hybrid neural-classical correction for adapting frozen TimesFM (200M parameters) to stock return prediction during the volatile opening trading hour. We compare two neural correction architectures - AttnCorrect (multi-head self-attention, approximately 471K parameters) and GatedLinear (low-rank bilinear projection with gating, approximately 49K parameters) - each augmented with Random Forest residual learning. Through systematic ablation across 10 major technology stocks (NVDA, MSFT, AAPL, GOOG, GOOGL, AMZN, META, AVGO, TSLA, NFLX) spanning 2 million data points, we reveal critical insights: (1) The hybrid neural-classical approach achieves 0.597 pooled correlation and 6.4x mean per-day correlation improvement over frozen TimesFM; (2) Classical residual learning (Random Forest) provides the largest single-component contribution, matching or exceeding the neural correction component; (3) Simpler neural architectures surprisingly outperform complex ones when classical residual learning is removed; (4) Self-attention provides the largest neural-only contribution. GatedLinear+RF achieves best overall performance with 9x fewer neural parameters than AttnCorrect+RF. We report three complementary correlation metrics - mean per-day, cross-day cumulative, and pooled - to provide a complete picture of predictive quality. Our results provide practical guidance: effective foundation model adaptation requires careful integration of neural and classical components, with classical methods playing a crucial complementary role.
