skills.rest
wer-inc
wer-incOFFICIAL
Explore all skills by this author →
GitHub

atft-research

Quantify Sharpe, RankIC, and hit ratio metrics across horizons and cohorts.

Updated Aug 28, 2025
Search Tags:#machine learning#investment#quantitative analysis#financial modeling#factor diagnostics#performance reporting#time series
One-click install
npx skills add https://github.com/wer-inc/gogooku3 --skill atft-research
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: atft-research
Source: https://github.com/wer-inc/gogooku3/tree/main/claude/skills/atft-research
Command: npx skills add https://github.com/wer-inc/gogooku3 --skill atft-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, matplotlib, seaborn, wandb, tensorboard, pytest, and includes scripts (resource) components.

What problem does it solve?

This Skill eliminates the tedious manual work of analyzing complex financial model outputs, making it hard to quickly understand performance, identify issues like factor drift, or generate stakeholder reports. It automates the quantification of model performance and diagnostics.

Core Features & Use Cases

  • Automated Performance Reporting: Generate comprehensive reports with key metrics like Sharpe, RankIC, and hit ratios across various horizons and cohorts.
  • Factor Diagnostics: Inspect feature contributions, leakage risks, and stability of graph-based factors to ensure model health and prevent unexpected behavior.
  • Use Case: Quickly validate a new model's output by generating a full research report and comparing it against a curated benchmark, ensuring it meets performance targets before deployment. This saves hours of manual data aggregation and visualization.

Quick Start

Example: Summarize a run and plot metrics

make research-baseline RUN=runs/<timestamp> python scripts/research/plot_metrics.py --run runs/<timestamp> --horizons 1 5 10 20

Frequently Asked Questions about atft-research

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

FAQPage Schema
How do I calculate Sharpe ratio, RankIC, and hit ratio for financial model outputs?▼

Sharpe ratio, RankIC, and hit ratio are performance metrics that quantify returns adjusted for risk, ranking correlation, and prediction accuracy. This Skill automates their calculation across time horizons and cohorts from your model predictions, eliminating manual computation and enabling quick performance validation before deployment.

How can I detect factor drift and leakage in machine learning financial models?▼

Factor drift occurs when feature importance or stability degrades over time; leakage occurs when training data inadvertently includes future information. This Skill inspects feature contributions and graph-based factor stability across runs, flagging risks and ensuring model health before production.

Can I automate generation of financial model performance reports with pandas and matplotlib?▼

Yes. This Skill automates report generation using pandas for data aggregation and matplotlib and seaborn for visualization, producing stakeholder-ready dashboards and notebooks that summarize performance metrics, factor diagnostics, and baseline comparisons without manual data wrangling.

What data inputs and artifacts do I need to analyze quantitative model performance?▼

You need model run artifacts (predictions.parquet), the full ML dataset (ml_dataset_latest_full.parquet), and feature metadata (dataset_features_detail.json). This Skill processes these inputs to produce comprehensive performance analysis, diagnostics, and compliance reports.

How do I compare a new financial model against a baseline to validate performance targets?▼

Baseline comparison runs your new model's metrics against a curated benchmark across horizons and cohorts, surfacing performance gaps or improvements. This Skill automates the comparison, saving hours of manual validation work before deployment decisions.

Can this work with time series financial data and multiple investment horizons?▼

Yes. The Skill is purpose-built for time series financial modeling, quantifying performance across multiple horizons (1, 5, 10, 20 periods) and market regimes, supporting regime segmentation and risk analysis across temporal dimensions.

Related Skills

Explore More Finance & Accounting Skills→
santoosaraujosantoosaraujo

ml-strategy

Generate predictive trading signals from OHLCV data using machine learning models.

Community
Advanced
WebWakaHubWebWakaHub
20

Test Results Analyzer

Analyze test execution results to identify failure patterns and release readiness risks.

Official
Advanced
afel6afel6

weights-and-biases

Track ML experiments and artifacts with Weights & Biases across PyTorch, TensorFlow, and HuggingFace.

Community
Intermediate
20YN0420YN04

alpha-zoo

Benchmark alpha factor libraries using IC and IR metrics.

Community
Advanced