Saved in:
Bibliographic Details
Main Authors: Lai, Tianyou, Yue, Wentao, Zhou, Jiayi, Hao, Chaoyuan, Chang, Lingke, Mao, Qingyu, Niu, Zhibo, Li, Qilei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.20869
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908905431367680
author Lai, Tianyou
Yue, Wentao
Zhou, Jiayi
Hao, Chaoyuan
Chang, Lingke
Mao, Qingyu
Niu, Zhibo
Li, Qilei
author_facet Lai, Tianyou
Yue, Wentao
Zhou, Jiayi
Hao, Chaoyuan
Chang, Lingke
Mao, Qingyu
Niu, Zhibo
Li, Qilei
contents Financial time-series forecasting in real-world high-frequency markets is often hindered by delayed or partially stale observations caused by asynchronous data acquisition and transmission latency. To better reflect such practical conditions, we investigate a simulated delay setting where a portion of historical signals is corrupted by a Zero-Order Hold (ZOH) mechanism, significantly increasing forecasting difficulty through stepwise stagnation artifacts. In this paper, we propose ReLaMix (Residual Latency-Aware Mixing Network), a lightweight extension of TimeMixer that integrates learnable bottleneck compression with residual refinement for robust signal recovery under delayed observations. ReLaMix explicitly suppresses redundancy from repeated stale values while preserving informative market dynamics via residual mixing enhancement. Experiments on a large-scale second-resolution PAXGUSDT benchmark demonstrate that ReLaMix consistently achieves state-of-the-art accuracy across multiple delay ratios and prediction horizons, outperforming strong mixer and Transformer baselines with substantially fewer parameters. Moreover, additional evaluations on BTCUSDT confirm the cross-asset generalization ability of the proposed framework. These results highlight the effectiveness of residual bottleneck mixing for high-frequency financial forecasting under realistic latency-induced staleness.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20869
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ReLaMix: Residual Latency-Aware Mixing for Delay-Robust Financial Time-Series Forecasting
Lai, Tianyou
Yue, Wentao
Zhou, Jiayi
Hao, Chaoyuan
Chang, Lingke
Mao, Qingyu
Niu, Zhibo
Li, Qilei
Artificial Intelligence
Machine Learning
Financial time-series forecasting in real-world high-frequency markets is often hindered by delayed or partially stale observations caused by asynchronous data acquisition and transmission latency. To better reflect such practical conditions, we investigate a simulated delay setting where a portion of historical signals is corrupted by a Zero-Order Hold (ZOH) mechanism, significantly increasing forecasting difficulty through stepwise stagnation artifacts. In this paper, we propose ReLaMix (Residual Latency-Aware Mixing Network), a lightweight extension of TimeMixer that integrates learnable bottleneck compression with residual refinement for robust signal recovery under delayed observations. ReLaMix explicitly suppresses redundancy from repeated stale values while preserving informative market dynamics via residual mixing enhancement. Experiments on a large-scale second-resolution PAXGUSDT benchmark demonstrate that ReLaMix consistently achieves state-of-the-art accuracy across multiple delay ratios and prediction horizons, outperforming strong mixer and Transformer baselines with substantially fewer parameters. Moreover, additional evaluations on BTCUSDT confirm the cross-asset generalization ability of the proposed framework. These results highlight the effectiveness of residual bottleneck mixing for high-frequency financial forecasting under realistic latency-induced staleness.
title ReLaMix: Residual Latency-Aware Mixing for Delay-Robust Financial Time-Series Forecasting
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2603.20869