The Physics of Price Discovery: Deconvolving Information, Volatility, and the Critical Breakdown of Signal during Retail Herding

Battle Count: 80

Highly relevant for quantitative trading. Provides actionable insights: (1) S_MC normalization superior to S_TV for institutional signal extraction; (2) regime-conditional price impact models necessary - constant-coefficient models are misspecified during herding; (3) size-dependent efficiency means cross-sectional strategies should account for herding intensity; (4) retail flow as contrarian signal for sophisticated traders; (5) Hawkes intensity as real-time warning indicator for trade timing; (6) negative convexity during herding periods affects execution strategy. Directly informs alpha signal construction, execution timing, and risk management.

Moderate-to-high complexity. Requires: (1) Tikhonov-regularized deconvolution with cross-validation for lambda selection; (2) Hawkes process MLE estimation with stability constraints; (3) Entropy production rate computation with permutation tests and block bootstrap; (4) Regime classification and conditional estimation; (5) Size-stratified analysis across quintiles; (6) Multiple robustness checks (regularization methods, thresholds, normalizations). Computational demands include 1000 bootstrap iterations, 500 block bootstrap resamples, 200 permutation shuffles. Requires investor-classified order flow data which is market-specific and not universally available.

Detailed methodology provided including regularization parameters (lambda=5.0), threshold choices (1.5 sigma), subsampling procedures (100 stocks, 5 iterations), and bootstrap protocols (1000 iterations). However, no code repository is mentioned, and the Korean Stock Exchange investor classification data may not be freely accessible. Robustness tests across regularization methods, thresholds, and normalizations are extensively documented.