Active Technological Similarity as a Predictor of Mutual Fund Performance

Summary

The study summary examines whether active stock selection that increases a mutual fund portfolio’s technological similarity predicts subsequent performance. Company patent distributions are compared using cosine similarity, and a fund’s active technological similarity (ATS) is the change in portfolio similarity relative to a no-trading portfolio. The proposed mechanism is that informed managers increase exposure to undervalued technological areas or to undervalued companies within those areas.

Using domestic actively managed equity funds, patent data, holdings, and returns, the study reports that higher ATS predicts stronger subsequent Carhart-adjusted returns. The relationship persists after controls and appears largely distinct from measures such as active share, industry concentration, and prior alpha; combined rankings can better separate future performance. The reported advantage is chiefly associated with trades that raise ATS. These are historical observational results from a defined sample period, not proof of causation or a guarantee of persistence, and the measure depends on patent and holdings data quality.

Key ideas

• ATS measures changes in a fund’s portfolio technological similarity attributable to trading.
• The study links increases in ATS with higher subsequent risk-adjusted mutual fund returns.
• ATS is reported to contain information largely distinct from prior alpha and common portfolio activity measures.
• Combining ATS rankings with other fund characteristics improves separation of future performance in the reported analysis.
• The performance association is concentrated in trades that increase technological similarity.

Tags

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