Paper: arXiv 2603.13638

Authors: Lucas A. Souza

Abstract

We introduce a performance-driven framework for constructing strictly causal forward-oriented observables in strongly non-stationary time series. The method combines a robustly normalized composite of heterogeneous indicators with a causally computed derivative component, yielding a local phase-leading effect that is amplified near regime transitions while remaining fully causal. A hysteresis-based decision functional maps the observable into discrete system states, with execution delayed by one step to preserve strict temporal ordering. Adaptation is achieved through a walk-forward scheme, in which model parameters are selected using rolling train–validation windows and subsequently applied out-of-sample. In this setting, the validation segment acts as an internal performance screen rather than as a statistical validation set, and no claims of generalization are inferred from it alone. The framework is evaluated on high-frequency financial time series as an experimentally accessible realization of a non-stationary complex system. Under a controlled zero-cost setting, the resulting dynamics exhibit a pronounced risk-reshaping effect, characterized by smoother trajectories and reduced drawdowns relative to direct exposure, and should be interpreted as an upper bound on achievable performance. These results illustrate how causal signal engineering can generate anticipatory structure in non-stationary systems without relying on non-causal information, explicit horizon labeling, or high-capacity predictive models.

Complexity vs Empirical Score

  • Math Complexity: 5.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a solid conceptual and dynamic signal construction framework aimed at solving non-stationarity without lookahead bias. While empirical backtesting under zero-cost conditions serves primarily as a proof-of-concept upper bound, the causal engineering methodology is novel and formal enough to demonstrate value.

Research Flowchart

  flowchart TD
    A[Research Goal: Causal Signal Engineering for Non-Stationary Financial Markets] --> B[Data Inputs: High-Frequency Financial Time Series]
    B --> C[Signal Engineering: Combine Normalized Indicators & Causal Derivative]
    C --> D[Decision Mechanism: Hysteresis Functional with 1-Step Delayed Execution]
    D --> E[Adaptive Process: Walk-Forward Optimization via Rolling Windows]
    E --> F[Outcomes: Reduced Drawdowns, Smoother Trajectories & Risk Reshaping]

Cited References (25)

  1. Duration Dependence in Stock Prices (2004) — link
  2. A simple framework for analysing bull and bear markets (2003) — link
  3. Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading Methods and Applications (1999)
  4. New Concepts in Technical Trading Systems (1978)
  5. The Moving Average Convergence–Divergence Trading Method: Advanced Techniques for Operating in Bull and Bear Markets (1979)
  6. Microstructure and Market Dynamics in Crypto Markets (2024) — link
  7. Arbitrage, contract design, and market structure in Bitcoin futures markets (2022) — link
  8. Informer In Algorithmic Investment Strategies on High Frequency Bitcoin Data — → 2503.18096
and 17 more
  1. Fact or friction: Jumps at ultra high frequency (2014) — link
  2. Backtest overfitting in the machine learning era: A comparison of out-of-sample testing methods in a synthetic controlled environment (2024) — link
  3. Detecting causality in non-stationary time series (2015) — link
  4. Identification of causal relationships in non-stationary time series (2022) — link
  5. Critical reflexivity in financial markets: a Hawkes process analysis (2013) — link
  6. Causal discovery from nonstationary time series (2025) — link
  7. Real-World Viability of Cointegration-Based Forex Pairs Trading Strategy with Walk-Forward Optimization (2025)
  8. Time Series Analysis: Forecasting and Control, 5th Edition, by George E. P. Box, Gwilym M. Jenkins, Gregory C. Reinsel and Greta M. Ljung, 2015. Published by John Wiley and Sons Inc., Hoboken, New Jersey, pp. 712. ISBN: 978-1-118-67502-1 (2016) — link
  9. Time Series Analysis (1994) — link
  10. DeepVol: volatility forecasting from high-frequency data with dilated causal convolutions (2024) — link
  11. Deep differentiable reinforcement learning and optimal trading (2022) — link
  12. Dynamic portfolio optimization across hidden market regimes (2017) — link
  13. Deep hedging (2019) — link
  14. Statistical arbitrage with optimal causal paths on high-frequency data of the S&P ~500 (2019) — link
  15. Forward-Oriented Causal Observables for Non-Stationary Financial Markets — → 2512.24621
  16. Practical Risk-Adjusted Performance Measurement (2022) — link
  17. Public API Documentation (2024)

2 of 25 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.