Risk attribution is a crucial step in advanced portfolio analysis. It moves beyond simply reporting the total risk of a portfolio (like its total Conditional Value at Risk) and allows managers to decompose that risk into granular, accountable components. This process identifies exactly which segments of the portfolio (e.g., asset classes, countries, sectors) are contributing the most to the overall downside risk.

The core analysis breaks down the portfolio's total risk into two distinct, yet related, measurements for each category: the 

(component risk). This framework is applied consistently across three key risk measures: Volatility, Value at Risk (VaR), and Conditional Value at Risk (CVaR).

The system provides attribution for three standard measures:

 of returns around the average. It captures both upside and downside deviation.

 over a specified time horizon at a given confidence level (e.g., 95% VaR means there is a 5% chance the loss will be greater than the VaR value). It is a threshold measure.

 Also known as Expected Shortfall (ES). It is the 

 in the tail of the distribution, specifically, the average loss that occurs 

 the VaR threshold. CVaR is considered superior for risk budgeting because it is a 

 that satisfies the property of subadditivity, which VaR generally does not. This property is what allows its total value to be perfectly broken down into additive component parts.

Comparison between Optimization Strategies

Risk Attribution

Category Risk (Stand-Alone)

MantaRisk Documentation

MantaRisk

Introduction

Getting Started (MantaAPI)

API Reference

Data Sources

MantaRisk API

MantaWealth for Addepar

Untitled