Portfolio Performance Attribution with Brinson and Factor Models

Summary

The document outlines three ways to explain portfolio results: style analysis using return regressions, benchmark-relative return decomposition, and multi-factor attribution. It describes the data each approach needs, including portfolio and benchmark returns and weights, plus stock-level factor exposures for factor analysis.

In the Brinson framework, active return is separated into allocation, security selection, and interaction effects. A factor model divides stock returns into common-factor and idiosyncratic components, then uses beginning-period portfolio exposures and estimated factor returns to attribute performance. Combining Brinson and factor methods can offer a more detailed view of return sources, while comparing attribution with risk exposures and intended portfolio preferences can help identify weak exposures or investigate sharp performance changes. The document presents these as analytical uses rather than empirical findings; it supplies no worked examples, performance results, or discussion of model assumptions and estimation limits.

Key ideas

• Style analysis uses return regressions to infer a portfolio’s investment style.
• Benchmark-relative attribution separates returns into allocation, selection, and interaction effects.
• Factor attribution decomposes returns into common-factor contributions and an idiosyncratic component.
• Attribution requires portfolio-specific inputs such as weights, returns, and factor exposures.
• Comparing realized attribution with intended portfolio preferences can help reveal ineffective exposures.

Tags

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