# Factor Library - Browse and Manage Factor Library

**Last Updated**: 2026-03-17
**Version**: 1.0.0

## How It Works

### Open the Factor Library

Navigate to the Factor Library page to access your centralized repository of alpha factors. The library contains all factors discovered through mining, imported from research, or manually created, organized in a searchable and sortable table.

### Browse and Sort Factors

View the complete list of factors with key metrics displayed in columns including factor name, IC, ICIR, annualized return, creation date, and source (Auto Discovery, From Report, or Optimize). Click column headers to sort by any metric for quick ranking.

### Search and Filter

Use the search bar to find factors by name or keyword. Apply filters to narrow results by performance range (e.g., IC > 0.03), source mode, creation date, or category. This helps you quickly locate high-quality factors from large libraries.

### Inspect Factor Details

Click on any factor to open its detail view showing the complete formula, Python implementation code, evaluation metrics (IC, ICIR, return, turnover), and historical performance chart. Review the factor's logic to understand its economic rationale.

### Evaluate Factor Correlations

Check correlation metrics between factors before combining them in a strategy. Highly correlated factors provide redundant signals, while low-correlation factors offer true diversification. The library displays pairwise correlation data for your factor set.

### Manage Factor Lifecycle

Activate, deactivate, rename, or delete factors from your library. Active factors are available for strategy integration, while deactivated factors are archived but preserved. Use tags and notes to organize factors by theme (momentum, value, quality, volatility).

### Apply Factors to Strategies

Select one or more factors from your library and integrate them into trading strategies via the Workflow Tabler. Combined factor portfolios can generate more robust signals than individual factors by blending multiple alpha sources.

> The Factor Library is your persistent repository for all discovered and curated alpha factors. Factor metrics are computed on the training dataset used during mining and should be re-validated on fresh data periodically.

## Tips & Best Practices

- Regularly audit your factor library by removing underperforming factors (IC near zero or negative ICIR) to keep the collection focused on high-quality signals

- When building multi-factor strategies, select 3-5 factors with low pairwise correlations (below 0.3) to maximize diversification benefit and reduce model instability

- Tag and categorize factors by their economic theme (momentum, mean-reversion, volatility, fundamental) to quickly assemble thematic factor portfolios

- Re-evaluate factor performance quarterly on recent data to detect factor decay; alpha signals can weaken as markets evolve and more participants adopt similar strategies

## Frequently Asked Questions

### What is an alpha factor and how does it differ from a trading signal?

An alpha factor is a quantitative formula that produces a cross-sectional score for assets, predicting relative future returns. A trading signal is a binary or directional decision (buy/sell/hold). Factors are continuous values that rank assets by expected return, and they become signals when combined with a threshold or portfolio construction rule.

### How should I interpret factor turnover?

Factor turnover measures how much the factor's rankings change between periods. High turnover means frequent rebalancing is needed, increasing transaction costs. Low turnover suggests stable rankings and lower trading costs. Aim for factors with meaningful IC that also have manageable turnover for practical implementation.

### Can I manually create or import factors into the library?

Factors are primarily added through the Factor Mining process (Auto Discovery, From Report, or Optimize modes). Factors that pass evaluation thresholds during mining are automatically saved to your library. You can manage and curate them from the library page.

### What does factor decay mean and how do I detect it?

Factor decay occurs when an alpha factor's predictive power diminishes over time, typically because the market has priced in the information the factor exploits. Detect it by comparing recent-period IC against historical IC. A sustained decline in rolling IC over 3-6 months signals potential decay, and the factor may need to be retired or re-optimized.

## Important Notes

> Factor performance metrics are based on historical evaluation data and may not persist in future periods. Alpha factors are subject to decay, crowding, and regime changes. Periodic re-evaluation on out-of-sample data is essential.

---

Source: https://stratcraft.ai/help/factor-library/