Using Black-Litterman to Add Analyst Forecasts to Multi-Factor Portfolios

Summary

This research summary compares two ways to incorporate analyst expectations into equity selection. A parallel approach blends forecasts with other factors or uses them jointly to predict returns. The proposed serial approach first builds a conventional multi-factor portfolio without analyst forecasts, then applies a Black-Litterman model outside that portfolio to revise stock weights using analyst expectations. The motivation is that analyst data often contain missing values, which can make direct factor blending difficult.

The report examines analyst target returns and target revenue growth. It says serial combinations outperformed parallel combinations and ordinary multi-factor portfolios in many cases. However, the summary provides no detailed performance figures or experimental design, and the underlying factor portfolios were not extensively optimized. It also emphasizes that results depend on how the base portfolio is constructed and which analyst forecast variables are used, so the finding is not a universal guarantee of improved performance.

Key ideas

• Analyst expectations can be blended directly with other factors or applied after a base portfolio is built.
• The serial method uses Black-Litterman to revise weights in a conventional multi-factor portfolio.
• Missing analyst data motivate keeping forecasts outside the initial factor construction.
• The report considers analyst target returns and target revenue growth.
• Reported comparative benefits depend on portfolio construction and forecast type.

Tags

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