| Engine | Home
 
RiskServers SA
Market Data Schema
Repository Guide
 
Keywords: Market Data Schema, Risk Factor Management, Time-Series storage. Stochastic Data.
Audience: Intermediate and Advanced users
   
Market Data Schema

Market Data Definition Tables: 

Foreword:

Proper Time-Series Data management must support different configurations.
This configuration depends largely on the amount and type of data you will be covering.

If you are managing multiple equity indices with their respective constituents, you might need to store series in multiple tables. In this case, the EQUITY_TABLE should hold additional tables that will include terms and conditions.
Considering today's computing power, we do however recommend you stick with single tables for equities and increase the swap space available for your database server. We only really recommend using multiple Equity tables if data import becomes a bottleneck.

This configuration is always determined by the number of Risk Factors multiplied by the number of historical points you will be importing.

Because of the Large number of vertices, curves are broken down into smaller classes, namely CURVE_import_govt Government Riskless Curves, CURVE_import_swap for Swap Curves, CURVE_import_mm for Money Market Instrument Risk Factors and CURVE_import_corp for Corporate (factors with specific Rating Classes).


The definition of what is used is always defined in the MARKET_DATA_DEFINITION table
The Market Data Definition table holds the basic definitions of Market Data:

  • Denomination Currency for Stored Currency Rates (Usually USD)
  • Groups of Tables for Equity and Commodity when multiple definitions are required
  • The Names of Curve Import Breakdown and Riskless and Spread Curve Vertex Time-Series storage
    • etc.

Market Data Schema Organization: 


The Market Data Schema holds the definition of Tables related to Market Risk Factor Data.
Tables in this group are designed to hold:
  • Risk Factor Definition per Asset Class.
  • Store Large Amounts of Snapped Time Series Data.
  • Store Computed Risk Factor Stochastic Values. (Added Value Data Management).
The Market Schema covers the entire spectrum of Market Data for Financial-Risk-Management analytics:
  • Automatic Risk Factor Import from external source.
  • Automatic Creation of new Risk Factors.
  • Market Data Definition Management.
  • Scrubbing and filtering policies applied to external data.
  • Time Series Price Management (Reordering, Computing Missing Points.
  • Export to external systems.
  • Compute Stochastic Variables: Volatilities, Correlations, Mean-Reversions
  • Stochastic Variable Storage.
  • Compute Volatility of Volatility, implied volatilities, correct non-definite multivariate data.

    For the sake of clarity, the Market Data Schema can be separated into three parts:
    1. The Risk Factor Import and Export Tables.
    2. The Risk Factor Definition Tables.
    3. The Risk Factor Price or Level Storage.

    Market Data Table Relationship

    1. The Risk The Risk Factor Import / Export Tables.

      Data import tables have three (somewhat) obvious characteristics:

      • All Market Data import tables have the _import suffix in their name.

        • For example the EQUITY table used to import Equity Data is named EQUITY_import.

        This feauture is actually used for all import tables.

      • _import tables used to import Market Data can be used
        1. Define Risk Factors
        2. Populate Time Series/ History of Prices.
        • _import tables have extra columns to Flag data for processing purposes.

      are always carry the _import suffix name.

      These tables Always have the following Pattern:
      "ASSET TYPE"_import.
      Where Asset Type is either CURRENCY, EQUITY, COMMODITY, CURVE, VOL, etc.
      Example:
      CURRENCY_import
      COMMODITY_Import
      EQUITY_import

    2. The Risk Factor Definition Tables  Risk Factor Vertices can be separated into two classes:
      • Risk Factors that have no Tenor Maturity. (ie. they are Spot or Cash)
        Typically:
        CURRENCY EQUITY
      • Risk Factors with Tenor Maturities:
        CURVE (INTERET RATES): either BASE (RISKLESS) or SPREAD
        COMMODITIES (Commodity Curves).
        IMPLIED VOLLATILITIES (Implied Volatility Surfaces or Tensors)
        FORWARD RATES Volatility and Correlations (Derived from Interest Rates Time Series)
        MEAN REVERSION FACTORS. (Derived from Interest Rates Time Series Multivariate Regression.)
    3. Market Risk Factors:
      This is where Observed Market Prices are Stored.
      Either all risk Factors are stored in one common table called
      MARKET_VERTEX. (not recommended)

      or each asset type has it's own storage, which are then aggregated when needed.
    4. MARKET_VERTEX_CCY (Currencies)
      MARKET_VERTEX_EQ (Equities)
      MARKET_VERTEX_COM (Commodities)
      MARKET_VERTEX_BASE (Base (Riskless Curves - Interest Rate Risk Factors)
      MARKET_VERTEX_SPREAD (Spread (Swap, Money Market, Corporate  Risk Factors).
      MARKET_VERTEX_IV (Implied Volatility Cells).
      Derived Vertices (as opposed to Observed Vertices) are also stored under this format, namely (MARKET_VERTEX_VOL and MARKET_VERTEX_CORRELATION)