Working Paper: CEPR ID: DP18129
Authors: Ron Kaniel; Zihan Lin; Markus Pelger; Stijn van Nieuwerburgh
Abstract: We show, using machine learning, that fund characteristics can consistently differentiate high from low-performing mutual funds, before and after fees. The outperformance persists for more than three years. Fund momentum and fund flow are the most important predictors of future risk-adjusted fund performance, while characteristics of the stocks that funds hold are not predictive. Returns of predictive long-short portfolios are higher following a period of high sentiment. Our estimation with neural networks enables us to uncover novel and substantial interaction effects between sentiment and both fund flow and fund momentum.
Keywords: mutual fund performance; fund flow; momentum; machine learning; sentiment; big data; neural networks
JEL Codes: G11; G12; G17; G23; C45
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| fund characteristics (G23) | mutual fund performance (G23) |
| fund momentum (G31) | mutual fund performance (G23) |
| fund flow (E50) | mutual fund performance (G23) |
| investor sentiment (G41) | mutual fund performance (G23) |
| fund flow + sentiment (E50) | mutual fund performance (G23) |
| fund momentum + sentiment (G41) | mutual fund performance (G23) |
| high sentiment (D64) | predictive long-short portfolios returns (G17) |
| predictive long-short portfolios returns (G17) | mutual fund performance (G23) |