居中驱动归一化收益：横截面收益预测中的价格偏移干扰

中文摘要 AI 辅助

基于原始日内柱的横截面收益预测对每个工具的价格水平敏感，在收益排序假设下，这是一种加性干扰。我们检验了去除这种偏移（而非重新缩放振幅或改变编码器）是否能解释在时点CSI 300五分钟面板上的收益提升。我们在有无RevIN归一化的情况下评估了八个参数匹配的编码器；一个无参数阶梯随后在所有字段和受限通道上分离了居中、仅缩放、居中、最后值引用、差分和标准化。居中驱动了可靠的效果，而仅缩放归一化并无帮助。所有八个配对效应均为正，并在原始秩IC、风格残差化后以及额外对短期反转残差化后均通过了Holm校正。在六个更强的编码器中，归一化IC为0.0830-0.0939，收益为0.0376-0.0567。仅价格标准化保留了全字段收益的93-101%。这些结果将主要效应归因于变换后的价格通道偏移去除，而非振幅缩放或编码器选择。

英文摘要

Cross-sectional return prediction from raw intraday bars is sensitive to each instrument price level, an additive nuisance under a return-ranking hypothesis. We test whether removing this offset, rather than rescaling amplitudes or changing the encoder, explains gains on a point-in-time CSI~300 five-minute panel. We evaluate eight parameter-matched encoders with and without RevIN normalization; a parameter-free ladder then separates identity, scale-only, centering, last-value referencing, differencing, and standardization across all fields and restricted channels. Centering drives the reliable effect, while scale-only normalization does not help. All eight paired effects are positive and survive Holm correction on raw rank IC, after style residualization, and after further residualizing on short-term reversal. Among six stronger encoders, gains of 0.0376-0.0567 exceed the 0.0109 spread of normalized IC (0.0830-0.0939). Price-only standardization retains 93--101% of the all-field gain. These results place the main effect in transformed price-channel offset removal rather than amplitude scaling or encoder choice.

发表机构

• Shanghai University（上海大学）
• School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
• The University of Sydney（悉尼大学）
• Qiuzhen College, Tsinghua University（清华大学邱耀学院）
• Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）

机构由 AI 辅助整理，请以论文原文为准。