​

Core Concepts

​

What is FRED?

​

Key Features

​

What is MCP?

​

MCP Benefits

• Standardized Interface: Consistent way to access external tools
• Security: Controlled access to resources
• Flexibility: Works with any MCP-compatible client
• Extensibility: Easy to add new capabilities

​

How FRED MCP Server Works

​

Architecture Components

1. MCP Client (e.g., Claude Desktop)
   • Sends tool requests
   • Receives structured data
   • Presents results to user
2. FRED MCP Server
   • Translates MCP requests to FRED API calls
   • Handles authentication
   • Processes and formats responses
   • Manages rate limiting
3. FRED API
   • Official Federal Reserve data source
   • RESTful API endpoints
   • JSON/XML responses

​

Data Organization

​

Series Structure

​

Category Hierarchy

Root Categories
├── Money, Banking & Finance
│   ├── Interest Rates
│   ├── Exchange Rates
│   └── Banking
├── Population, Employment & Labor Markets
│   ├── Unemployment Rate
│   ├── Employment
│   └── Labor Force
└── Production & Business Activity
    ├── GDP
    ├── Industrial Production
    └── Business Surveys

​

Data Sources

• U.S. Government Agencies
  • Bureau of Labor Statistics (BLS)
  • Bureau of Economic Analysis (BEA)
  • Census Bureau
  • Federal Reserve Board
• International Organizations
  • World Bank
  • OECD
  • IMF
  • Bank for International Settlements
• Academic Institutions
  • University research centers
  • Economic research organizations

​

Time Series Concepts

​

Frequency

​

Seasonal Adjustment

• SA (Seasonally Adjusted): Seasonal effects removed
• NSA (Not Seasonally Adjusted): Raw data with seasonal patterns
• SAAR (Seasonally Adjusted Annual Rate): Adjusted and annualized

​

Transformations

​

Data Quality

​

Vintage Data

• Real-time data: Current values
• Vintage data: Values as they existed on specific dates
• Revisions: Updates to previously released data

​

Data Updates

• High-frequency: Updated daily or weekly
• Monthly indicators: Released on schedule (e.g., first Friday)
• Quarterly data: Released with lag (e.g., GDP one month after quarter end)
• Annual series: Updated yearly

​

Missing Data

• Some series have incomplete histories
• Holidays and weekends create gaps in daily data
• Series may be discontinued or replaced

​

API Concepts

​

Rate Limiting

• 120 requests/minute: General limit
• 40 requests/minute: For series observations
• Daily limit: Varies by API key tier

• Request queuing
• Exponential backoff
• Retry logic

​

Pagination

{
  "limit": 100,    // Items per page
  "offset": 200,   // Skip first 200 items
  "total": 1500    // Total available items
}

​

Error Handling

​

Search and Discovery

​

Search Methods

1. Text Search: Find series by keywords
2. Tag Search: Filter by categories and attributes
3. Browse: Navigate category hierarchy
4. Release-based: Find series by data release

​

Tag System

• Geographic: usa, state, county, msa
• Frequency: daily, weekly, monthly
• Topic: gdp, inflation, employment
• Source: bls, bea, federal reserve

​

Popularity Ranking

• Based on access frequency
• Helps identify commonly-used indicators
• Range from 1-100

​

Best Practices

​

Efficient Queries

​

Data Interpretation

​

Common Patterns

​

Economic Analysis Pattern

1. Discover relevant series using search/browse
2. Retrieve data with appropriate transformations
3. Compare across time periods or series
4. Analyze trends and relationships

​

Monitoring Pattern

1. Identify key indicators to track
2. Schedule regular data retrieval
3. Alert on significant changes
4. Report findings

​

Research Pattern

1. Browse categories for comprehensive coverage
2. Search for specific relationships
3. Retrieve long historical periods
4. Transform for statistical analysis

​

Next Steps