Survivorship Bias in Emerging Market Small-Cap Indices: Evidence from India's NIFTY Smallcap 250

Battle Count: 79

Extremely high relevance. The paper directly addresses a fundamental methodological flaw in quantitative strategy backtesting: using current index constituents as proxies for historical universes. The 23.3% return overstatement and 9.1% Sharpe ratio inflation are large enough to flip investment allocation decisions. The findings apply directly to any quant researcher, hedge fund, or proprietary trading desk evaluating small-cap strategies in India or similar emerging markets. The paper provides both the diagnostic (magnitude of bias) and the prescription (use complete historical universes, discount returns by 20-25%). The methodology for reconstructing historical universes from publicly available data is directly actionable for practitioners.

Moderate complexity. The core methodology (ranking stocks by traded value, selecting ranks 151-400, constructing equal-weight portfolios) is algorithmically straightforward. However, the data processing pipeline requires handling 2,459 files with varying formats across years, standardizing schemas, filtering equity securities, and managing date parsing across multiple formats. The bootstrap significance testing and decomposition analysis add additional steps. No ML model training is required. The main challenge is data acquisition and cleaning rather than algorithmic complexity. A competent quantitative analyst could replicate this in 2-4 weeks with Python/pandas.

Data is publicly available via NSE bhavcopy files (Samco Securities archive). Methodology is described in extensive detail including processing pipeline, reconstruction algorithm, validation procedures, and mathematical formulas. The 100% validation against current constituents provides strong confidence. However, no code repository is provided, and the exact processing scripts are not shared. The 85-90% historical reconstruction accuracy estimate introduces some uncertainty that cannot be fully verified without official historical constituent lists.