survivorship-bias

Eliminate survivorship bias in historical backtests using point-in-time membership snapshots.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/jefrnc/quant-llm-skills --skill survivorship-bias
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: survivorship-bias
Source: https://github.com/jefrnc/quant-llm-skills/tree/main/skills/survivorship-bias
Command: npx skills add https://github.com/jefrnc/quant-llm-skills --skill survivorship-bias

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents inflated backtests and misleading performance comparisons caused by accidentally using today’s universe, today’s index membership, or delisted/successor ticker data when simulating historical results.

Core Features & Use Cases

  • Point-in-time universe reconstruction: Rebuilds the membership at each rebalance/query date using the correct at-that-time snapshot rather than today’s surviving constituents.
  • Delisting-aware pricing & exit treatment: Detects common vendor/data pitfalls (successor mapping, adjusted-feed artifacts, reverse-split-delist phantoms) and enforces realistic handling (e.g., mark-to-zero unless a realizable value is documented).
  • Index/fund/leaderboard integrity: Ensures rankings and track records include dropouts at their last-reported value instead of silently removing them.
  • Small-cap and FPI high-risk focus: Targets the patterns where survivorship bias is most severe, such as reverse-split-then-delist, ATM-into-delisting, SPAC merger discontinuities, and Reg SHO threshold residency.

Quick Start

Use the skill whenever you’re assembling a backtest universe for a historical date range and need the analysis to include delisted names as-of their correct membership and exit timing.

Frequently Asked Questions about survivorship-bias

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent survivorship bias in historical trading backtests?

To prevent survivorship bias in backtesting, reconstruct point-in-time universes using membership snapshots at each rebalance date, rather than relying on today's surviving constituents or current index membership.

Why does my small-cap backtest show inflated returns?

Small-cap backtests show inflated returns when survivorship bias silently removes delisted names, reverse-split phantoms, or SPAC merger discontinuities, requiring delisting-aware exit treatment to enforce realistic mark-to-zero values.

What is point-in-time universe reconstruction for index membership?

Point-in-time universe reconstruction rebuilds the correct at-that-time snapshot for each query date, ensuring historical performance comparisons include dropouts at their last-reported value instead of silently removing them.

How do I handle delisted tickers and vendor successor mapping in historical data?

Handle delisted tickers and vendor successor mapping by detecting adjusted-price artifacts and enforcing realistic exit treatment, marking delisted securities to zero unless a realizable value is explicitly documented.

Do I need point-in-time data validation for fund track record calculations?

You need point-in-time data validation for fund track record calculations to ensure rankings and leaderboards include dropouts at their last-reported value, preventing misleading performance comparisons across time.