Selecting Bond Funds with Factor Attribution and Residual Alpha

Summary

This research describes a bond-fund selection method built around return attribution. It expands the Campisi framework—which separates income, government-rate, credit-spread, and security-selection effects—with convertible-bond and monetary-policy effects. Seven index-derived style factors represent yield-curve level, slope and convexity, credit, default, convertible bonds, and monetary conditions. Regressing fund net-value returns on these factors estimates exposures and residual alpha; funds are ranked by that alpha.

The reported backtest uses rolling six-month estimation windows, selects the top tenth of eligible funds at quarter ends, weights them equally, and holds them for three months. The document reports results over roughly five years, including excess returns versus an equally weighted fund universe, and says selection remained effective after accounting for convertible-bond exposure. These are historical findings, not guarantees: the source explicitly warns that past performance may not recur, the model may fail, and market shocks could cause large fund fluctuations. The supplied text summarizes the study but does not include the full methodological details or data needed to independently reproduce its results.

Key ideas

• The method decomposes bond-fund returns into style-factor exposures and residual alpha.
• Seven factors cover rate-curve characteristics, credit and default risk, convertible bonds, and monetary conditions.
• Funds are ranked by regression alpha using a rolling lookback, then selected and held on a quarterly schedule.
• The reported backtest finds excess performance both with and without convertible-bond effects.
• Historical results may not persist, and model failure or market shocks can impair outcomes.

Tags

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