Gary Antonacci has just posted a new research paper on absolute momentum. I encourage everyone to check it out. Some of his previous research served as a catalyst for the Dual ETF Momentum tracker on Scott's Investments.
The current signals as of Wednesday's close for the Dual ETF Momentum portfolio is below. As a reminder, the portfolio is not intended as an optimal portfolio or the ideal mix of asset classes, but does show how to put relative and absolute momentum into action:
Return data courtesy of Finviz
|Equity||Symbol||1 Year % Total Returns||Average of Quarterly/Half/Full Year % Returns||Signal based on 1 year returns||Signal based on average returns|
|Credit Risk||Symbol||1 Year % Total Returns||Average of Quarterly/Half/Full Year % Returns||Signal based on 1 year returns||Signal based on average returns|
|High Yield Bond||HYG||12.17||6.57||Invested||Invested|
|Interm Credit Bond||CIU||5.59||2.46|
|Real-Estate Risk||Symbol||1 Year % Total Returns||Average of Quarterly/Half/Full Year % Returns||Signal based on 1 year returns||Signal based on average returns|
|Economic Stress||Symbol||1 Year % Total Returns||Average of Quarterly/Half/Full Year % Returns||Signal based on 1 year returns||Signal based on average returns|
I decided to backtest variants of the absolute momentum strategy Antonacci discusses in his newest paper. I used ETFReplay.com for the tests and created mutual fund portfolios (as opposed to ETFs) due to their longer trading histories. For commodities, I used the GSCI Index.
The basic premise of absolute momentum is that a security must outperform a risk-free asset (such as short-term treasuries or "cash") over the look-back period (i.e. the past 6, 10, 12 month, etc., returns).
The test below was rebalanced monthly, a 12-month look-back period was used and the Vanguard Short-Term Treasury Fund (MUTF:VFISX) was used as the cash filter. In other words, if a security in the portfolio had not outperformed VFISX over the past 12 months then the security was sold and replaced with cash for one month.
The Parity Portfolio was comprised of the following securities and was created to closely (but not precisely) mimic the "Parity Portfolio" used in Antonacci's absolute momentum paper:
|GTY||S&P GSCI Commodity (Index)|
|VFICX||Vanguard Intermediate-Term Investment-Grade Fund|
|VGSIX||Vanguard REIT Index Fund|
|VTSMX||Vanguard Total Stock Market Index Fund|
|VUSTX||Vanguard Long-Term Treasury Fund|
I tested a variety of portfolios from 2003 to April 8th, 2013. The Vanguard S&P 500 Index Fund (MUTF:VFINX) was tested both as a buy-and-hold and with a 12-month absolute momentum strategy. The Parity Portfolio was tested as a buy-and-hold (with an annual rebalance) and with a 12-month absolute momentum strategy. For comparison purposes I also tested the historical returns of the Vanguard Balanced Index Fund (MUTF:VBINX).
The test results:
|2003-Present||VFINX (S&P 500)||VFINX w/ abs Momentum||VBINX (60/40), Buy & Hold||Parity Portfolio (rebalanced annually)||Parity Portfolio w/ abs Momentum|
Using 12-month absolute momentum improved risk adjusted returns for both the Parity Portfolio and VFINX. Maximum drawdown and volatility was also significantly reduced. Total return for the annual rebalance Parity Portfolio was slightly higher, but at higher volatility and with a -40.15% drawdown versus the -11.90% drawdown using absolute momentum on the same portfolio.
Food for thought!