Testing Moving Average Trading Strategies on ETFs
44 Pages Posted: 12 Mar 2018
Date Written: March 12, 2018
This paper tests the technical trading rule of moving average (MA) in a long-only portfolio using exchange traded funds (ETFs). We also propose a quasi-intraday version of the MA strategy (QUIMA) that allows investors to trade immediately upon observing MA crossover signals. We find that 1) this QUIMA strategy outperforms the traditional version of the MA strategy that only trades at the close of a trading day, when the long-term MA lag length is not too long, 2) the documented profitability of MA strategy on indices is greatly reduced on ETFs, mainly due to more frequent and larger opening gaps on ETF prices than those on indices, and 3) relative to the buy-and-hold strategy, MA strategies have lower return, but better risk-adjusted performance measures such as the CAPM alpha. In addition, we find that among various long-term MA lengths, the 10-day MA turns out to be overly exploited by investors as its performance is significantly lower than those of surrounding MA lengths. Overall, our findings indicate that profitability of the MA trading rule reduces on tradable ETFs than on non-tradable indices.
Keywords: Moving Average, Technical Trading Rules, Data-Snooping Bias, Exchange Traded Funds
JEL Classification: G11, G14, D83
Suggested Citation: Suggested Citation