Skip to content
All library documents

Stratified Convertible Bond Investing with Segment-Specific Factors

Article BigQuant

Summary

This study builds a convertible-bond strategy that separates bonds into equity-like, balanced, and bond-like groups using parity premium relative to bond value. It models each bond as a floor component plus option time value, then examines where factor returns arise within that decomposition. The preferred signals differ by group: conversion premium and an amplitude pattern for equity-like bonds, conversion premium and bond-versus-stock amplitude spread for balanced bonds, and amplitude spread plus current yield for bond-like bonds.

At each month end, the method standardizes and equally combines relevant factors within each group, selects the highest-scoring bonds, and holds an equal-weight portfolio from the next month's first trading day. The document reports historical performance for the full stratified portfolio and for a version excluding equity-like bonds. Those figures come from a backtest that excludes transaction and market-impact costs; the analysis also warns that elevated valuations and sharp reversals can hurt equity-like bonds, especially those with high recent turnover and returns. Historical results therefore provide limited evidence about future performance.

Key ideas

  • The method classifies convertible bonds into three groups based on parity premium.
  • It decomposes bond value into a floor component and option time value to study factor returns.
  • Signal selection varies by group rather than applying one universal factor set.
  • The portfolio ranks bonds monthly and equally weights selected issues within each group.
  • The reported backtest excludes trading frictions and is vulnerable to changing market conditions.

Tags

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