---
title: "校准赌注于过去：金融时间序列预测的后训练量化研究"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.12259v1"
permalink: "https://feed.astralor.com/items/itm_d3843502cf55b79be49c794f"
published: "2026-08-12T17:02:06.000Z"
score: 75
channel: "tech"
tags: ["arXiv","S&P 500","后训练量化","激活校准","金融时间序列预测"]
---

# 校准赌注于过去：金融时间序列预测的后训练量化研究

## AI 点评

> 这篇论文的价值在于把量化部署中常被忽视的校准环节，从‘技术细节’提升为‘一级决策’。对做金融模型工程化的人来说，4位量化不再是单纯的压缩问题，而是需要和市场波动特性绑定的建模问题。尤其值得注意的是，百分位校准能挽回大部分损失，这意味着许多现有部署可能只需改一行代码就能显著改善性能。

## 速览

研究表明，在金融时间序列预测中，后训练量化（PTQ）的激活校准对8位量化影响甚微，但在4位量化时成为预测性能的主要决定因素。基于S&P 500横截面波动率预测，覆盖7种神经网络架构、8年测试期（2018-2025）及560个训练模型，默认绝对最大值校准在4位静态量化下导致受影响架构的全精度平均信息系数损失11-62%，而百分位校准可恢复53-94%的退化。激活范围需随市场状态调整，对可靠4位PTQ至关重要。

## 关键点

- 在4位量化下，激活校准成为预测性能的主要决定因素，而8位量化下影响甚微。
- 默认abs-max校准导致信息系数下降11-62%，百分位校准可恢复大部分退化。
- 最优激活范围随市场波动性变化，历史校准窗口可能无法覆盖未来极端波动。
- 对于退化严重的场景，8位激活或仅权重量化是更稳健的部署选择。

## 全文

Financial forecasting models are typically developed in full precision, yet production deployment often requires low-precision inference to reduce memory and computational cost. Post-training quantization (PTQ) enables such deployment without retraining. However, reliable activation quantization requires calibration: activation ranges are estimated from historical data before deployment and then remain fixed during future inference. The importance of this deployment choice for financial forecasting remains poorly understood. We present a systematic study of activation calibration for PTQ in cross-sectional volatility forecasting on the S&P 500. Our evaluation covers seven representative neural architectures, eight walk-forward test years (2018-2025), and 560 trained models. We find that activation calibration has little effect at 8 bits but becomes the primary determinant of predictive performance at 4 bits. Under default absolute-maximum (abs-max) calibration, static 4-bit quantization of both weights and activations removes 11-62% of the full-precision mean information coefficient in affected architectures. Replacing abs-max with percentile calibration recovers 53-94% of this degradation in the four most affected architectures. The preferred activation range also varies across market periods. Narrow ranges improve resolution under typical market conditions but lose part of their advantage when test-period market dispersion exceeds the calibration history. These findings show that activation calibration is a first-class deployment decision for reliable 4-bit PTQ in financial forecasting. When substantial degradation remains, 8-bit activations or weight-only 4-bit quantization provide more robust deployment choices.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_d3843502cf55b79be49c794f