Replication Data for: Asymmetric Hurst-Gated LPPL Detection: Evidence from Segmented Institutional/Retail Index Futures

Description

This dataset contains LPPL calibration results, rolling Hurst estimates, backtest outputs, and Python analysis scripts for the Finance Research Letters manuscript. Ten CSV files cover parameter comparison, bootstrap validation, Hurst ablation, slippage sensitivity, and placebo policy tests. Seven Python scripts reproduce all tables and figures.

Steps to reproduce

1. Install Python 3.12+ with: pip install numpy pandas scipy duckdb matplotlib 2. Run run_backtest_ih_im.py to reproduce Table 1 (all 6 strategies) 3. Run hurst_ablation.py to reproduce Hurst method comparison 4. Run e01_e02_analysis.py for parameter comparison figures 5. Run robustness_checks.py for lagged Hurst, leverage, and placebo tests 6. All output files save to ./output/ directory

Categories

Licence