Paper: arXiv 2601.05716

Authors: Sungwoo Kang

Abstract

Most empirical microstructure research assumes that order flow–return parameters are constant, yet these relationships shift substantially across market regimes. Combining adaptive Kalman filtering, Markov-switching regime identification, and asymmetric response estimation, we characterize regime-dependent investor behavior in the Korean stock market during 2020–2024 using daily transaction data disaggregated by investor type. Three principal findings emerge: foreign investor predictive power increases several-fold during crisis periods relative to bull markets; individual investors chase momentum asymmetrically, reacting far more strongly to positive than to negative shocks; and independent information-theoretic validation corroborates both patterns. Rigorous out-of-sample testing reveals that these in-sample regularities do not generalize reliably, underscoring the need for proper validation methodology in microstructure research.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 6.0/10 (capped from 7.5: transaction costs only partially modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a solid econometric framework combining adaptive state-space filtering, Markov-switching, and asymmetric response modeling applied to Korean market order flow. It demonstrates commendable empirical rigor by highlighting out-of-sample generalization failure rather than over-selling in-sample results.

Research Flowchart

  flowchart TD
    A[Research Goal: Analyze Regime-Dependent Order Flow & Return Dynamics] --> B[Data: Korean Stock Market Daily Transactions 2020-2024 by Investor Type]
    B --> C[Methodology: Adaptive Kalman Filtering & Markov-Switching Regimes]
    C --> D[Computational Analysis: Asymmetric Response & Information-Theoretic Validation]
    D --> E[In-Sample Findings: Foreign Predictive Power Spikes in Crises; Retail Chases Upside Momentum]
    D --> F[Out-of-Sample Testing: In-Sample Patterns Fail to Generalize Reliably]