Skip to content
All library documents

Portfolio Performance Attribution with Brinson and Multifactor Models

Article BigQuant

Summary

The document explains how performance attribution can reveal the sources of a portfolio’s returns, even when portfolios have similar styles and overall results. It outlines three approaches: style analysis, which uses portfolio and benchmark returns to infer investment style; benchmark-relative return decomposition, which separates allocation and security selection; and multifactor analysis, which attributes results to exposures such as value or growth factors.

For Brinson attribution, active return is divided into allocation, selection, and interaction effects. A multifactor model separates returns explained by common factors from idiosyncratic returns, using beginning-of-period portfolio exposures and estimated factor returns. The document argues that combining return and risk models can help identify inefficient exposures and compare realized performance with portfolio-construction preferences. It provides a conceptual overview rather than empirical results, and does not specify model variants, estimation choices, or implementation details; the attribution conclusions depend on the inputs and model assumptions.

Key ideas

  • Performance attribution helps distinguish return sources that aggregate performance figures can obscure.
  • Style analysis infers investment style by regressing portfolio returns against benchmark returns.
  • Benchmark-relative attribution separates active return into allocation, selection, and interaction effects.
  • Multifactor attribution links portfolio returns to factor exposures and separates common-factor returns from specific returns.
  • Comparing realized attribution with intended exposures can inform investment process review.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.