skills.rest
HKUST-QUANT-SOCIETY
HKUST-QUANT-SOCIETYOFFICIAL
Explore all skills by this author →
GitHub

factor-evaluation

Evaluates quantitative factors through the canonical QuantEvaluator and records contracts, provenance, and artifacts.

Updated Jun 27, 2026
Search Tags:#quantitative finance#contract validation#factor evaluation#quantcode#quant evaluator#ic ir metrics#factor panel
One-click install
npx skills add https://github.com/HKUST-QUANT-SOCIETY/quantcode --skill factor-evaluation-hkust-quant-society
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: factor-evaluation
Source: https://github.com/HKUST-QUANT-SOCIETY/quantcode/tree/main/.opencode/groups/factor/skills/factor-evaluation
Command: npx skills add https://github.com/HKUST-QUANT-SOCIETY/quantcode --skill factor-evaluation-hkust-quant-society

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Quantitative researchers need a single, trustworthy way to evaluate factors without duplicated or inconsistent metrics. This Skill routes all factor evaluation through the canonical QuantEvaluator, ensuring IC, IR, turnover, layering, and label metrics come from one authoritative source with full provenance tracking. ## Core Features & Use Cases - Canonical Evaluation Routing: Directs FactorPanel or FactorSpec inputs to QuantEvaluator, and routes missing factor values through DataAccess and FactorEngine first. - Contract and Status Validation: Verifies FactorSpec/FactorPanel contracts and surfaces UNAVAILABLE, PARTIAL, MOCK, PROXY, or STAGING statuses honestly instead of claiming production success. - Artifact and Provenance Recording: Displays component-returned artifacts on success and records contract, source, and version information for auditability. - Use Case: A factor group researcher has a new FactorPanel and needs standardized IC/IR metrics; the Skill validates the contract, calls quant_evaluator, and returns the official artifact without computing proxy metrics. ## Quick Start Evaluate my current FactorPanel with the canonical QuantEvaluator and show me the resulting artifact and status.

Frequently Asked Questions about factor-evaluation

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

FAQPage Schema
How do I evaluate a factor with QuantEvaluator in QuantCode?▼

Provide a valid FactorPanel or FactorSpec, and the Skill validates the contract then calls the quant_evaluator component. On SUCCEEDED status it displays the returned artifact; other statuses are shown as-is without claiming production success.

What if I have no factor values computed yet?▼

The Skill routes the workflow through DataAccess and then FactorEngine to produce factor values before invoking QuantEvaluator. It never computes its own proxy or mock IC, IR, or turnover metrics.

Does factor evaluation support structure or parameter search?▼

No, structure and parameter search is delegated to the FactorOptimizer component instead. This Skill only handles canonical evaluation of existing FactorPanel or FactorSpec inputs.

Why does my factor evaluation show PARTIAL or MOCK status?▼

These statuses come directly from the unified ComponentCallResult returned by QuantEvaluator. The Skill displays them verbatim so you know the evaluation is not a full production result, rather than substituting proxy metrics.

Can the skill write evaluation results to shared factor assets?▼

Real shared asset writes only go through publishing tools connected to the exact merge approval and receipt flow. If the legacy adapter is not connected, the Skill reports that explicitly instead of bypassing approval.

Related Skills

Explore More Data & Analytics Skills→
prinzevalprinzeval

factor-research

Evaluate factor signal predictive power across instruments using IC/IR metrics.

Community
Advanced
quantskillsquantskills
61

skill-quant-factor-skill-factory

Batch-generate and validate framework-neutral OHLCV quant factor Skills on real market data.

Official
Advanced
understudylabsunderstudylabs
10

compare-model-sweep

Evaluate candidate models across a fixed evaluation to identify the Pareto frontier.

Official
Advanced
tlapidus-rgbtlapidus-rgb

pricing-modeler

Model NitroSales pricing and ROI scenarios using canonical CONOCIMIENTO_PRODUCTO/PRECIOS.md data.

Community
Intermediate