The Low-volatility Anomaly and the Adaptive Multi-Factor Model
40 Pages Posted: 4 May 2021 Last revised: 3 Nov 2021
Date Written: April 25, 2021
Abstract
The paper provides a new explanation of the low-volatility anomaly. We use the Adaptive Multi-Factor (AMF) model estimated by the Groupwise Interpretable Basis Selection (GIBS) algorithm to find those basis assets significantly related to low and high volatility portfolios. These two portfolios load on very different factors, indicating that volatility is not an independent risk, but that it's related to existing risk factors. The out-performance of the low-volatility portfolio is due to the (equilibrium) performance of these loaded risk factors. The practical insight is that the long and short legs of a portfolio are with different risks and need to be modeled separately. Our methodology is applicable to any long-short anomaly but we focus on the low-volatility anomaly since it is formed explicitly on the risk characteristic rather than on embedded risks of other anomalies. The AMF model outperforms the Fama-French 5-factor model significantly both in-sample and out-of-sample.
Keywords: Low-volatility anomaly, AMF model, GIBS algorithm, high-dimensional statistics, machine learning, False Discovery Rate.
JEL Classification: C10, G10
Suggested Citation: Suggested Citation