Improving Time-Series Momentum Strategies: The Role of Volatility Estimators and Trading Signals
UBS AG; Imperial College Business School
Imperial College Business School; University of Oxford, Oxford-Man Institute of Quantitative Finance
July 30, 2013
The aim of this paper is to examine the effect of risk-weighting and of the choice of trading signal on the performance of time-series momentum strategies using a broad dataset of 75 futures contracts over the period 1974-2013. Time-series momentum strategies have received increased attention after they provided again, as in previous business cycle downturns, impressive diversification benefits during the recent financial crisis in 2008. Motivated by recent asset pricing literature that examines the effect of frictions on asset prices and the link between portfolio volatility and turnover, we highlight the effect of the choice of volatility estimator and trading signal on turnover and performance of time-series momentum strategies. We find that by increasing the efficiency of volatility estimation using estimators with desirable theoretical properties, such as range-based estimators, the net of transaction costs performance improves, but the effect on turnover is relatively small compared to that of the trading signal. Momentum trading signals generated by fitting a linear trend on the asset price path maximise the out-of-sample performance by reducing portfolio turnover by about two thirds, hence dominating other momentum trading signals commonly used in the literature.
Number of Pages in PDF File: 35
Keywords: Trend-following, Momentum, Managed Futures, Volatility Estimation, Trading Signal, Transaction Costs
JEL Classification: D23, E3, G14working papers series
Date posted: September 2, 2012 ; Last revised: July 31, 2013
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo1 in 0.422 seconds