Paper: arXiv 2609.29887

Authors: Yi-Chen Liu, Chung-Han Hsieh

Abstract

This paper investigates cost-sensitive online window size selection for portfolio management under changing market conditions. Specifically, we propose a two-level framework that constructs portfolios using candidate window sizes and dynamically aggregates them through online learning. By treating candidate window sizes as ``experts,’’ we dynamically update their aggregation weights using turnover-inclusive losses. Moreover, we derive finite-horizon cost-sensitive tracking-regret bounds that account for turnover of the aggregated portfolio, with static regret as a special case. Under bounded losses and cost rates, suitably tuned Fixed Share achieves asymptotically no tracking regret for sublinear switching budgets, with Hedge covering the static case.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Lab Rats — theoretically deep, empirically untested

Why this score: The paper presents a highly mathematical framework for cost-sensitive online window selection with strong theoretical guarantees. While the theoretical rigor is high, the excerpt provided does not detail any empirical validation or backtesting, which is crucial for quant finance. The novelty lies in integrating turnover costs into regret bounds for dynamic window selection.

Research Flowchart

  flowchart TD
    A[Research Goal: Cost-Sensitive Online Window Size Selection for Portfolio Management] --> B{Key Methodology: Two-Level Framework};
    B -- Level 1: Portfolio Construction --> C[Candidate Window Sizes as "Experts"];
    B -- Level 2: Dynamic Aggregation --> D[Online Learning: Update Aggregation Weights with Turnover-Inclusive Losses];
    C & D --> E[Computational Process: Derive Finite-Horizon Cost-Sensitive Tracking-Regret Bounds];
    E --> F{Key Findings: Asymptotically No Tracking Regret (Fixed Share), Static Case Covered (Hedge)};

Cited References (76)

  1. On Unified Adaptive Portfolio Management — → 2307.03391
  2. Literature review: Machine learning techniques applied to financial market prediction (2019)
  3. On-Line Portfolio Selection with Moving Average Reversion (2012) — link
  4. Approximation to Bayes Risk in Repeated Play (1957)
  5. Prediction, Learning, and Games (2006)
  6. Tests for Parameter Instability and Structural Change with Unknown Change Point (1993)
  7. Empirical Asset Pricing via Machine Learning (2020)
  8. On the optimality of the Hedge algorithm in the stochastic regime (2019)
and 68 more
  1. Portfolio Selection (1952)
  2. Foundations of Portfolio Theory (1991)
  3. Convex optimization (2004)
  4. Online convex optimization and no-regret learning: Algorithms, guarantees and applications (2018) — link
  5. CVXPY: A Python-Embedded Modeling Language for Convex Optimization (2016)
  6. An analog of the minimax theorem for vector payoffs. (1956)
  7. Blackwell Approachability and No-Regret Learning are Equivalent (2011)
  8. Tracking a Small Set of Experts by Mixing Past Posteriors (2001) — link
  9. Dynamic Regret Reduces to Kernelized Static Regret (2025) — link
  10. Learning with kernels: support vector machines, regularization, optimization, and beyond (2002)
  11. Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation (1982)
  12. Online Learning: A Modern Introduction Using Convex Optimization (2026) — link
  13. Does the Stock Market Overreact? (1985)
  14. Adaptive sliding window normalization (2025) — link
  15. Learning from Time-Changing Data with Adaptive Windowing (2007) — link
  16. Detecting Concept Drift with Support Vector Machines (2000)
  17. Experience with a learning personal assistant (1994) — link
  18. Adaptive information filtering: Learning in the presence of concept drifts (1998)
  19. The Weighted Majority Algorithm (1994) — link
  20. Big Data: Deep Learning for financial sentiment analysis (2018)
  21. Machine learning in finance: the case of deep learning for option pricing (2017)
  22. Advances in financial machine learning (2018)
  23. Identifying Optimal Window Size Configurations for Big Data Time Series Forecasting (2024) — link
  24. Determining the impact of window length on time series forecasting using deep learning (2019) — link
  25. Forecasting price movements using technical indicators: Investigating the impact of varying input window length (2017) — link
  26. Selection of Market Window Size in Portfolio Strategies (2018) — link
  27. Machine learning in the Chinese stock market (2022) — link
  28. Machine learning in finance (2020)
  29. International financial adjustment (2007)
  30. Out-of-Sample Forecast Tests Robust to the Choice of Window Size (2012)
  31. Out-of-Sample Exchange Rate Predictability with Taylor Rule Fundamentals (2009)
  32. Why has US inflation become harder to forecast? (2007)
  33. Selection of estimation window in the presence of breaks (2007)
  34. Forecast Combination Across Estimation Windows (2011) — link
  35. Optimal forecasts in the presence of structural breaks (2013)
  36. Learning in the presence of concept drift and hidden contexts (1996)
  37. Fractal Market Analysis: Applying Chaos Theory to Investment and Economics (1994)
  38. Rolling window selection for out-of-sample forecasting with time-varying parameters (2017)
  39. MLP-Based Learnable Window Size for Bitcoin Price Prediction (2022)
  40. Adaptive sliding window normalization (2025)
  41. Time-Varying Window Length for Correlation Forecasts (2017)
  42. Topic: No-Regret and Coarse Correlated Equilibria (2020)
  43. Risk-Sensitive Online Learning (2006)
  44. Efficient learning algorithms for changing environments (2009) — link
  45. Strongly Adaptive Online Learning (2015)
  46. Online Portfolio Selection: A survey (2014) — link
  47. Technical Note - A Robust Perspective on Transaction Costs in Portfolio Optimization (2018)
  48. Trading costs (2018)
  49. Measuring and modeling execution cost and risk (2006)
  50. A Closer Look at Adaptive Regret (2016)
  51. Equilibria under Dynamic Benchmark Consistency in Non-Stationary Multi-Agent Systems (2025) — link
  52. Tracking the Best Expert (1998)
  53. A New Interpretation of Information Rate (1956)
  54. Universal portfolios (1991)
  55. On-Line Portfolio Selection using Multiplicative Updates (1998)
  56. Universal Portfolio Selection (1998)
  57. Stochastic Nonstationary Optimization for Finding Universal Portfolios (2000)
  58. PAMR: Passive Aggressive Mean Reversion Strategy for Portfolio Selection (2012)
  59. Confidence weighted mean reversion strategy for online portfolio selection (2013)
  60. Optimal Portfolio Selection and Dynamic Benchmark Tracking (2005)
  61. On-line Portfolio Selection Using Stochastic Programming (2003)
  62. Mirror Descent Meets Fixed Share (and Feels No Regret) (2012)
  63. Tracking the Best Linear Predictor (2001)
  64. Relative Loss bounds for Multidimensional Regression Problems (1997)
  65. The Relaxation Method of Finding the Common Point of Convex Sets and Its Application to the Solution of Problems in Convex Programming (1967)
  66. On Information and Sufficiency (1951)
  67. Convex Optimization: Algorithms and Complexity (2015)
  68. Investment Science (2013)

1 of 76 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.