Latest · 0.1.0 Nov 20, 2025
PyPI file Logistic-bounded IPO risk scoring framework.
Latest v0.1.0
Quant-grade IPO risk scoring, without the black box.
IPO Risk Score maps IPO characteristics into a transparent, logistic-bounded score. Built for researchers, quants and operators who need interpretable risk signals in a clean Python API.
Score range
0 – 100
Logistic-bounded output
Validation
Input guards
Ranges, enums, null checks
Latency
< 10 ms
Pure Python, no network calls
Score preview
UPX-style IPO
Illustrative scoring from the working paper to show how feature stacks roll into the final number.
Risk score
98.94 / 100
Attractiveness
1.06 / 100
Drivers
- Micro float / liquidity drag
- Non-Big4 auditor + mid-tier UW
- Emerging-market exposure
Signals
Features stay monotonic: higher risk inputs always push the score up, avoiding directional surprises.
pip
Install
Install ipo-risk-score
Ship the scorer into notebooks, batch jobs or production services with a single command.
pip install ipo-risk-score
Latest version: 0.1.0 · Logistic-bounded IPO risk scoring framework.
Release signal
Pulled from PyPI in real time
Latest version
0.1.0
Nov 20, 2025
Logistic-bounded IPO risk scoring framework.
Auto-updates after each publish to PyPI.
Pipeline
How the scorer assembles risk
Data quality comes first, then monotonic features, then a bounded logistic transform. Each piece is isolated so you can iterate on signals without changing the contract.
Structured IPO intake
Typed IPO inputs enforce numeric ranges, enums and clean strings before any feature is derived.
Feature stacks
Liquidity, valuation, quality and geography live in isolated modules so each axis can evolve independently.
Logistic scoring
A weighted linear combination flows through a stable logistic transform, returning a risk score in [0, 100].
Capabilities
Built for readability and control
Modular stacks, guarded inputs and a bounded score make it simple to slot into research workflows or production systems.
Model
Bounded, interpretable output
0–100 scale with monotonic weights. Thresholds are easy to calibrate for gating, monitoring or alerts.
Governance
Audit-friendly pipeline
Strict validation, explicit feature weights and deterministic transformations make every score reproducible.
Architecture
Modular by design
Feature stacks are decoupled; extend valuation or quality signals without rewriting the core scorer.
Workflow
Research ready
Docstring-rich API suited for notebooks, batch backtests or realtime scoring inside Python workflows.
Release log
Live version history
Updates pull directly from PyPI so the site stays current whenever a new ipo_risk_score package ships.
Release feed
Pulled live from PyPI
Working paper
Logistic-Bounded Risk Scoring Framework for IPOs
A detailed walk-through of the model, assumptions, feature definitions and case studies.
Repository
Open-source implementation
Transparent, research-friendly Python package. Contributions and issues are welcome.