skills.rest
kofttlcc
kofttlccCOMMUNITY
Explore all skills by this author →
GitHub

advanced-feature-engineering

Apply fractional differentiation and rolling window standardization to financial time series data.

14|3|Updated Jan 17, 2026
Search Tags:#feature engineering#time series#normalization#stationarity#financial machine learning#fractional differentiation#look-ahead bias
One-click install
npx skills add https://github.com/kofttlcc/quant-test --skill advanced-feature-engineering
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: advanced-feature-engineering
Source: https://github.com/kofttlcc/quant-test/tree/main/.agent/skills/advanced_feature_engineering/quant-feature-eng
Command: npx skills add https://github.com/kofttlcc/quant-test --skill advanced-feature-engineering

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, numpy, statsmodels, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill generates high-quality features for financial machine learning models by addressing data non-stationarity and preventing look-ahead bias during standardization.

Core Features & Use Cases

  • Fractional Differentiation: Stabilizes time series data while preserving historical memory, crucial for financial modeling.
  • Look-ahead Bias Free Rolling Normalization: Ensures that feature scaling at any point in time only uses past data, maintaining data integrity.
  • Use Case: Prepare stock price data for a predictive model by applying fractional differentiation to achieve stationarity and then standardizing it using a rolling window to avoid look-ahead bias.

Quick Start

Use the advanced-feature-engineering skill to generate features for the provided financial time series data.

Frequently Asked Questions about advanced-feature-engineering

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

FAQPage Schema
How do I make financial time series data stationary without losing historical memory?▼

Rolling window standardization prevents look-ahead bias in feature scaling by ensuring that calculations at any point only use past data. This maintains time-series data integrity for financial modeling.

How do I normalize time series features without introducing look-ahead bias?▼

Rolling window standardization prevents look-ahead bias in feature scaling by ensuring that calculations at any point only use past data. This maintains time-series data integrity for financial modeling.

What is the best way to prepare stock price data for predictive machine learning models?▼

Statistical tests like the ADF (Augmented Dickey-Fuller) test are used to verify time series stationarity after applying fractional differentiation. This ensures the transformed financial data meets modeling requirements.

Does this feature engineering approach require specific Python dependencies?▼

You should avoid standardizing financial features with global statistics or future-looking rolling windows, as this introduces look-ahead bias. Strict past-data-only rolling calculations are required to preserve data integrity.

Related Skills

Explore More Data & Analytics Skills→
0xZKnw0xZKnw

seasonal

Generate trading signals from historical month-of-year and day-of-week effects.

Community
Intermediate
ssmleossmleo

bigquery-ai-ml

Run AI and machine learning functions for BigQuery SQL queries.

Community
Advanced
ebrahim-saniebrahim-sani

quant-statistics

Perform statistical tests and models for quantitative finance using Python libraries.

Community
Intermediate
RorySullivan1RorySullivan1

quant-code-review

Identify numerical defects in quantitative-finance Python code causing wrong results.

Community
Advanced